Publications

Read our published work with NEC, partner organizations and across our five departments: data science and system security, integrated systems, machine learning, media analytics, and optical networking and sensing.

Automated Negotiation and Multimodal Time-Series Forecasting for Efficient Procurement

May 29, 2026/The 25th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2026)

Procurement is a key function in supply chain management that involves acquiring goods and services to meet organizational needs. Efficient procurement is crucial for minimizing costs, ensuring timely delivery, and maintaining quality standards. This paper explores the integration of automated negotiation

Open-SAT: LLM-Guided Query Embedding Refinement for Open-Vocabulary Object Retrieval in Satellite Imagery

May 15, 2026/arXiv

In satellite applications, user queries often take the form of open-ended natural language, extending beyond a fixed set of predefined categories. This open-vocabulary nature poses significant challenges for retrieving relevant image tiles, as the retrieval system must generalize to a wide range of unseen

Closed-Form Statistical Modeling of PDL-Induced SNR Margins for Reliable Optical Networks

May 12, 2026/30th International Conference on Optical Network Design and Modelling (ONDM 2026)

We develop closed-form formulas for PDL-induced SNR margins using solutions based on central limit theorem. Experimental validations confirm accurate and conservative performancepredictions, enabling precise quality of transmission assessment and margin-aware design in optical networks.

RunAgent: Interpreting Natural-Language Plans with Constraint-Guided Execution (IEEE)

May 10, 2026/IEEE Conference on Artificial Intelligence 2026 (IEEE CAI 2026)

Humans solve problems by executing targeted plans, yet large language models (LLMs) remain unreliable for structured workflow execution. We propose RunAgent, a multiagent plan execution platform that interprets natural-language plans while enforcing stepwise execution through constraints and rubrics.

Event Classification by Physics-Informed Inpainting for Distributed Multichannel Acoustic Sensor with Partially Degraded Channels

May 4, 2026/2026 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2026)

Distributed multichannel acoustic sensing (DMAS) enables large-scale sound event classification (SEC), but performance drops when many channels are degraded and when sensor layouts at test time differ from training layouts. We propose a learning-free, physics-informed inpainting frontend based on reverse

Learning to Tune OpticalWANs: A Field Deployment of Noise Models in Optical Networks

May 4, 2026/23rd USENIX Symposium on Networked Systems Design and Implementation (NSDI '26)

Accurately modeling optical signal transmission is critical foroptimizing network performance, particularly in large-scalefiber optic networks operated by Internet Service Providers.In this work, we develop a Gaussian Noise model for a NewYork state ISPs optical backbone. Our model accounts for allmajor

Mix-Clap: Adaptive Fusion of Knowledge-Distilled Audio Embeddings for Noise-Aware Audio-Language Models

May 4, 2026/2026 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2026)

Real-world deployment requires sound event and acoustic scene classification systems to remain reliable in noisy, diverse environments on resource-constrained devices. Although contrastive language-audio pretraining (CLAP) models with Transformer-based audio encoders achieve strong zero-shot performance,

PhyCo: Learning Controllable Physical Priors for Generative Motion

May 4, 2026/arXiv

Modern video diffusion models excel at appearance synthesis but still struggle with physical consistency: objects drift, collisions lack realistic rebound, and material responses seldom match their underlying properties. We present PhyCo, a framework that introduces continuous, interpretable, and physically

GNPy as a Benchmark for Open and Disaggregated Optical Networks

May 1, 2026/Journal of Optical Communications and Networking

The evolution toward open and partially disaggregated optical networks has introduced new, to our knowledge,requirements on how transmission performance is evaluated and compared across technologies, vendors, and deployment scenarios. In this context, sound benchmarking practices are essential to ensure

Solving Inverse Problems via a Score-Based Prior: An Approximation-Free Posterior Sampling Approach

May 1, 2026

Diffusion models (DMs) have proven to be effective in modeling high-dimensional distributions, leading to their widespread adoption for representing complex priors in Bayesian inverse problems (BIPs). However, current DM-based posterior sampling methods proposed for solving common BIPs rely on heuristic

RunAgent: Interpreting Natural-Language Plans with Constraint-Guided Execution (arXiv)

April 28, 2026/arXiv

Humans solve problems by executing targeted plans, yet large language models (LLMs) remain unreliable for structured workflow execution. We propose RunAgent, a multi-agent plan execution platform that interprets natural-language plans while enforcing stepwise execution through constraints and rubrics.

Quantitative Bounds for Length Generalization in Transformers

April 23, 2026

We study the problem of length generalization (LG) in transformers: the ability of a model trained on shorter sequences to maintain performance when evaluated on much longer, previously unseen inputs. Prior work by Huang et al. (2024) established that transformers eventually achieve length generalization

Agentic Placement of Microservices on the Computing Continuum

April 19, 2026/The Seventeenth International Conference on Cloud Computing, GRIDs, and Virtualization (Cloud Computing 2026) - special Track (Hyper-CC)

Deploying microservices across the computing continuum (edgecloud) requires placement decisions that adapt to workload variation and heterogeneous infrastructure, yet existing solutions often rely on static policies or opaque heuristics. We present Bellona a system for reliable and auditable Large

Learning to Route: A Rule-Driven Agent Framework for Hybrid-Source Retrieval-Augmented Generation

April 12, 2026/The 2026 ACM Web Conference (WWW 2026)

Large Language Models (LLMs) have shown remarkable performance on general Question Answering (QA), yet they often struggle in domain-specific scenarios where accurate and up-to-date information is required. Retrieval-Augmented Generation (RAG) addresses this limitation by enriching LLMs with external

Uncertainty-Guided Latent Diagnostic Trajectory Learning for Sequential Clinical Diagnosis

April 6, 2026/https://arxiv.org

Clinical diagnosis requires sequential evidence acquisition under uncertainty. However, most Large Language Model (LLM) based diagnostic systems assume fully observed patient information and therefore do not explicitly model how clinical evidence should be sequentially acquired over time. Even when diagnosis

Leveraging Deployed Telecom Cables for Distributed Fiber Sensing Topologies and Applications

April 1, 2026/Journal of Optical Communications and Networking

Distributed fiber optic sensing (DFOS) has emerged as a promising technology for wide-area monitoring by utilizing existing telecom cables as large-scale sensing media. This paper explores three sensing modalities, backscattering-based sensing, forward-transmission-based sensing, and hybrid sensing,

HorizonWeaver: Generalizable Multi-Level Semantic Editing for Driving Scenes

April 1, 2026/https://arxiv.org

Ensuring safety in autonomous driving requires scalable generation of realistic, controllable driving scenes beyond what real-world testing provides. Yet existing instruction guided image editors, trained on object-centric or artistic data, struggle with dense, safety-critical driving layouts. We propose

Uncertainty-Aware Knowledge Distillation for Multimodal Large Language Models

April 1, 2026/https://arxiv.org

Knowledge distillation establishes a learning paradigm that leverages both data supervision and teacher guidance. However, determining the optimal balance between learning from data and learning from the teacher is challenging, as some samples may be noisy while others are subject to teacher uncertainty.

Decoding Time Series with LLMs: A Multi-Agent Framework for Cross-Domain Annotation

March 29, 2026/The 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026)

Time series data is ubiquitous across various domains, including manufacturing, finance, and healthcare. High-quality annotations are essential for effectively understanding time series and facilitating downstream tasks. However, obtaining such annotations is challenging, particularly in mission-critical

DeepSieve: Information Sieving via LLM-as-a-Knowledge-Router

March 29, 2026/19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026)

Large Language Models (LLMs) excel at many reasoning tasks but struggle with knowledge-intensive queries due to their inability to dynamically access up-to-date or domain-specific information. Retrieval-Augmented Generation (RAG) has emerged as a promising solution, enabling LLMs to ground their responses

Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement

March 29, 2026/The 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026)

Automatically extracting workflows as procedural graphs from natural language is promising yet underexplored, demanding both structural validity and logical alignment. While recent large language models (LLMs) show potential for procedural graph extraction, they often produce ill-formed structures or

Interpretability and Implicit Model Semantics in Biomedicine and Deep Learning

March 23, 2026/Nature Machine Intelligence

We introduce a framework to analyse interpretability in deep learning, by drawing on a formal notion of model semantics from the philosophy of science. We argue that interpretability is only one aspect of a models semantics and illustrate our framework with examples from biomedicine.

Agnostic QoT Probing via Receiver-Side ASE Loading in a Production Metro for Transparent Datacenter Exchange

March 15, 2026/OFC 2026

We demonstrate agnostic QoT probing for datacenter exchange in a metro network via receiver-side ASE loading. Knowing BER telemetry and the progressive ASEload, the device estimates GSNR, enabling IPoWDM operations and digital-twin calibration.

Field study on phase and polarization dynamics of deployed anti-resonant hollow core fiber cable for vibration sensing

March 15, 2026/OFC 2026

We report the first field study of the phase and polarization dynamics of deployed antiresonant hollow core fiber cable in a data center interconnect for real-world vibration sensing,revealing enhanced phase sensitivity and significantly faster polarization angular rate compared with standard single

Frequency-Division Multiplexed Time-Interleaved Phase-OTDR with Nested Phase References

March 15, 2026/OFC 2026

We propose a method to compensate the phase offset between samples from different tributaries in time-interleaved phase OTDR using nested phase reference channels. We demonstrate our method for a four-span bidirectional link with high-loss loopback.

Mobile Orbital Domain-based Hierarchical Routing in Satellite Networks

March 15, 2026/OFC 2026

We propose a mobile orbital domain-based hierarchical routing scheme which addresses the challenges posed by constant satellite movement and the resulting dynamicnetwork topology, thus significantly improving the routing scalability and efficiency in satellite networks.

Distilling Offline Action Detection Models into Real-Time Streaming Models

March 6, 2026/The IEEE/CVF Winter Conference on Applications of Computer Vision 2026, Tucson, Arizona

Vision Transformers (ViTs) have achieved state-of-the-art performance in offline video action detection, but their reliance on processing fixed-size clips with full spatio-temporal attention makes them computationally expensive and ill-suited for real-time streaming applications due to massive computational

Image-Specific Adaptation of Transformer Encoders for Compute-Efficient Segmentation

March 6, 2026/5th Workshop on Image/Video/Audio Quality Assessment in Computer Vision, VLM and Diffusion Model in conjunction with WACV 2026

Vision transformer-based models bring significant improvements for image segmentation tasks. Although these architectures offer powerful capabilities irrespective of specific segmentation tasks, their use of computational resources can be taxing on deployed devices. One way to overcome this challenge

Logical Guidance for the Exact Composition of Diffusion Models

March 1, 2026/arXiv

We propose LOGDIFF (Logical Guidance for the Exact Composition of Diffusion Models), a guidance framework for diffusion models that enables principled constrained generation with complex logical expressions at inference time. We study when exact score-based guidance for complex logical formulas can

HorizonForge: Driving Scene Editing with Any Trajectories and Any Vehicles

February 23, 2026/arXiv

Controllable driving scene generation is critical for realistic and scalable autonomous driving simulation, yet existing approaches struggle to jointly achieve photorealism and precise control. We introduce HorizonForge, a unified framework that reconstructs scenes as editable Gaussian Splats and Meshes,

Manhole Localization and Condition Diagnostics in Telecom Networks Using Distributed Acoustic and Temperature Sensing

February 1, 2026/Journal of Lightwave Technology

We present methods and field trial results demonstrating an integrated distributed acoustic sensing (DAS) and distributed temperature sensing (DTS) system for manhole localization, condition diagnostics, and anomaly detection in pre-deployed telecommunication fiber networks. The proposed system leverages

Field Trial of High-Sensitivity Forward-Transmission Sensing for Real-World Event Detection Over Live Urban Fiber Networks

February 1, 2026/Journal of Lightwave Technology

Vibration sensing based on forward transmission is an emerging topic for network protection and environmental monitoring, especially in long-haul submarine cables and urban fiber networks. However, previous field trials of this approach have mainly focused on localizing strong events under controlled

MARLIN: Multi-Agent Reinforcement Learning for Incremental DAG Discovery

January 27, 2026/40th AAAI Conference on Artificial Intelligence (AAAI-26)

Uncovering causal structures from observational data is crucial for understanding complex systems and making informed decisions. While reinforcement learning (RL) has shown promise in identifying these structures in the form of a directed acyclic graph (DAG), existing methods often lack efficiency, making

Brownian Bridge Augmented Surrogate Simulation and Injection Planning for Geological CO2 Storage

January 22, 2026/The 40th Annual AAAI Conference on Artificial Intelligence (AAAI-26)

Geological CO2 storage (GCS) involves injecting captured CO2 into deep sub-surface formations to support climate goals. The effective management of GCS relies on adaptive injection planning to dynamically control injection rates and well pressures to balance both storage safety and efficiency. Prior

Advances in Fiber Sensing

January 20, 2026/SPIE Photonics West

In this talk, we will present recent technological advances in fiber sensing applications with long monitoring distances orextending multiple fiber spans. In forward-transmission-based sensing, adaptive beamforming techniques weredemonstrated to achieve multi-event vibration sensing in environments with

Object-Aware 4D Human Motion Generation

January 5, 2026/arXiv

Recent advances in video diffusion models have enabled the generation of high-quality videos. However, these videos still suffer from unrealistic deformations, semantic violations, and physical inconsistencies that are largely rooted in the absence of 3D physical priors. To address these challenges,

EditGRPO: Reinforcement Learning with Post-Rollout Edits for Clinically Accurate Chest X-Ray Report Generation

December 20, 2025/2025 IJCNLP-AACL International Joint Conference on Natural Language Processing & Asia-Pacific Chapter of the Association for Computational Linguistics

Radiology report generation requires advanced medical image analysis, effective temporal reasoning, and accurate text generation. Although recent innovations, particularly multimodal large language models, have shown improved performance, their supervised fine-tuning (SFT) objective is not explicitly

Visual Alignment of Medical Vision-Language Models for Grounded Radiology Report Generation

December 18, 2025/arXiv

Radiology Report Generation (RRG) is a critical step toward automating healthcare workflows, facilitating accurate patient assessments, and reducing the workload of medical professionals. Despite recent progress in Large Medical Vision-Language Models (Med-VLMs), generating radiology reports that are

Optical Link Tomography: First Field Trial and 4D Extension

December 15, 2025/Journal of Lightwave Technology

Optical link tomography (OLT) is a rapidly evolving field that allows the multi-span, end-to-end visualization of optical power along fiber links in multiple dimensions from network endpoints, solely by processing signals received at coherent receivers. This paper has two objectives: (1) to report the

Online Multi-modal Root Cause Identification in Microservice Systems

December 11, 2025/2025 IEEE International Conference on Big Data

Root Cause Analysis (RCA) is essential for pinpointing the root causes of failures in microservice systems. Traditional data-driven RCA methods are typically limited to offline applications due to high computational demands, and existing online RCA methods handle only single-modal data, overlooking complex

Human Texts Are Outliers: Detecting LLM-generated Texts via Out-of-distribution Detection

December 7, 2025/The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025)

The rapid advancement of large language models (LLMs) such as ChatGPT, DeepSeek, and Claude has significantly increased the presence of AI-generated text in digital communication. This trend has heightened the need for reliable detection methods to distinguish between human-authored and machine-generated

Multi-Modal View Enhanced Large Vision Models for Long-Term Time Series Forecasting

December 7, 2025/The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025)

Time series, typically represented as numerical sequences, can also be transformed into images and texts, offering multi-modal views (MMVs) of the same underlying signal. These MMVs can reveal complementary patterns and enable the use of powerful pre-trained large models, such as large vision models

SolverLLM: Leveraging Test-Time Scaling for Optimization Problem via LLM-Guided Search

December 7, 2025/The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025)

Large Language Models (LLMs) offer promising capabilities for tackling complex reasoning tasks, including optimization problems. However, existing methods either rely on prompt engineering, which leads to poor generalization across problem types, or require costly supervised training. We introduce SolverLLM,

TimeXL: Explainable Multi-modal Time Series Prediction with LLM-in-the-Loop

December 7, 2025/The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025)

Time series analysis provides essential insights for real-world system dynamics and informs downstream decision-making, yet most existing methods often overlook the rich contextual signals present in auxiliary modalities. To bridge this gap, we introduce TimeXL, a multi-modal prediction framework that

TacTool: Tactical Tool usage in Agentic AI Systems

December 5, 2025/2025 IEEE International Conference on Agentic AI (ICA)

Large language models (LLMs) are becoming the centerpiece in the design and deployment of Agentic artificial intelligence (AI) systems. AI agents typically have (a) reasoning ability to analyze and think through the given task, (b) context/memory to remember things in the short-term and long-term, and

To Err Is Human: Systematic Quantification of Errors in Published AI Papers via LLM Analysis

December 4, 2025/arXiv

How many mistakes do published AI papers contain? Peer-reviewed publications form the foundation upon which new research and knowledge are built. Errors that persist in the literature can propagate unnoticed, creating confusion in follow-up studies and complicating reproducibility. The accelerating pace

iFinder: Structured Zero-Shot Vision-Based LLM Grounding for Dash-Cam Video Reasoning

December 2, 2025/The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025)

Grounding large language models (LLMs) in domain-specific tasks like post-hoc dash-cam driving video analysis is challenging due to their general-purpose training and lack of structured inductive biases. As vision is often the sole modality available for such analysis (i.e. no LiDAR, GPS, etc.), existing

Uni-LoRA: One Vector is All You Need

December 2, 2025/The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025)

Low-Rank Adaptation (LoRA) has become the de facto parameter-efficient fine-tuning (PEFT) method for large language models (LLMs) by constraining weight updates to low-rank matrices. Recent works such as Tied-LoRA, VeRA, and VB-LoRA push efficiency further by introducing additional constraints to reduce

Integrated Optical-to-Optical Gain in a Silicon Photonic Modulator Neuron

December 1, 2025/Optica

Silicon photonic neural networks can achieve higher throughputs and lower latencies than digital electronic alternatives.However, recently reported implementations of such networks have lacked integrated signal gain, instead utilizingoff-chip amplifiers or co-processors to complete the signal processing

DISC: Dynamic Decomposition Improves LLM Inference Scaling

December 1, 2025/The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025)

Inference scaling methods for LLMs often rely on decomposing problems into steps (or groups of tokens), followed by sampling and selecting the best next steps. However, these steps and their sizes are often predetermined or manually designed based on domain knowledge. We propose dynamic decomposition,

xTime: Extreme Event Prediction with Hierarchical Knowledge Distillation and Expert Fusion

November 15, 2025/25th IEEE International Conference on Data Mining (IEEE ICDM 2025)

Extreme events frequently occur in real-world time series and often carry significant practical implications. In domains such as climate and healthcare, these events, such as floods, heatwaves, or acute medical episodes, can lead to serious consequences. Accurate forecasting of such events is therefore

Correlation-aware Online Change Point Detection

November 14, 2025/The 34th ACM International Conference on Information and Knowledge Management (CIKM 2025)

Change point detection aims to identify abrupt shifts occurring at multiple points within a data sequence. This task becomes particularly challenging in the online setting, where different types of change can occur, including shifts in both the marginal and joint distributions of the data. In this paper,

Quantitative Bounds for Length Generalization in Transformers

November 10, 2025/The Fourteenth International Conference on Learning Representations (ICLR 2026)

We study the problem of length generalization (LG) in transformers: the ability of a model trained on shorter sequences to maintain performance when evaluated on much longer, previously unseen inputs. Prior work by Huang et al. (2025) established that transformers eventually achieve length generalization

Neuromorphic Photonics-Enabled Near-Field RF Sensing with Residual Signal Recovery and Classification

November 9, 2025/IEEE Photonics Conference (IPC 2025)

We present near-field radio-frequency (RF) sensing using microwave photonic canceler (MPC) for residual signal recovery and neuromorphic photonic recurrent neural network (PRNN)chip and FPGA hardware to implement machine learning for high-bandwidth and low-latency classification.

Scalable Photonic Neurons for High-speed Automatic Modulation Classification

November 9, 2025/IEEE Photonics Conference (IPC 2025)

Automatic modulation classification (AMC) is becoming increasingly critical in the context of growing demands for ultra-wideband, low-latency signal intelligence in 5G/6G systems, with photonics addressing the bandwidth and real-time adaptability limitations faced by traditional radio-frequency (RF)

Sound Event Classification meets Data Assimilation with Distributed Fiber-Optic Sensing

October 30, 2025/DCASE 2025 Workshop

Distributed Fiber-Optic Sensing (DFOS) is a promising technique for large-scale acoustic monitoring. However, its wide variation in installation environments and sensor characteristics causes spatial heterogeneity. This heterogeneity makes it difficult to collect representative training data. It also

SlideCraft: Context-aware Slides Generation Agent

October 21, 2025/The 23rd IEEE International Conference on Pervasive Intelligence and Computing (PICom 2025)

Creating effective slide presentations requires adapting both content and structure to match the communication context e.g. whether the presentation is for summarizing to executives, or reporting progress to research supervisors. In research and enterprise environments, this need for context-sensitive

TalentScout: Multimodal AI-Driven Expert Finding in Organizations

October 21, 2025/The 23rd IEEE International Conference on Pervasive Intelligence and Computing (PICom 2025)

Identifying subject-matter experts within organizations remains a challenging task due to the scale, heterogeneity, and unstructured nature of enterprise knowledge assets. We present TalentScout, an AI-driven expert identification system that constructs a unified, skill-centric knowledge graph by ingesting

AutoScape: Geometry-Consistent Long-Horizon Scene Generation

October 19, 2025/ICCV 2025

This paper proposes AutoScape, a long-horizon driving scene generation framework. At its core is a novel RGB-D diffusion model that iteratively generates sparse, geometrically consistent keyframes, serving as reliable anchors for the scenes appearance and geometry. To maintain long-range geometric consistency,

DWIM: Towards Tool-aware Visual Reasoning via Discrepancy-aware Workflow Generation & Instruct-Masking Tuning

October 19, 2025/ICCV 2025

Visual reasoning (VR), which is crucial in many fields for enabling human-like visual understanding, remains highly challenging. Recently, compositional visual reasoning approaches, which leverage the reasoning abilities of large language models (LLMs) with integrated tools to solve problems, have shown

LANGTRAJ: Diffusion Model and Dataset for Language-Conditioned Trajectory Simulation

October 19, 2025/ICCV 2025

Evaluating autonomous vehicles with controllability enables scalable testing in counterfactual or structured settings, enhancing both efficiency and safety. We introduce LangTraj, a language-conditioned scene-diffusion model that simulates the joint behavior of all agents in traffic scenarios. By conditioning

Mapillary Vistas Validation for Fine-Grained Traffic Signs: A Benchmark Revealing Vision-Language Model Limitations

October 19, 2025/The 4th DataCV Workshop and Challenge at ICCV 2025

Obtaining high-quality fine-grained annotations for traffic signs is critical for accurate and safe decision-making in autonomous driving. Widely used datasets, such as Mapillary, often provide only coarse-grained labels without distinguishing semantically important types such as stop signs or speed

Emerging Integrated Photonic Technologies Leveraging Multimaterial Integration for AI and Datacenter Applications

October 6, 2025/Optics Express

Since the inception of integrated photonics, multimaterial integration has served as a primary avenue for new technology innovations. Now, with an ever-increasing demand for integrated photonics as a platform for both high-performance links from/within datacenters and AI acceleration, multimaterial integration

THAT: Token-wise High-frequency Augmentation Transformer for Hyperspectral Pansharpening

October 5, 2025/IEEE SMC 2025

Transformer-based methods have demonstrated strong potential in hyperspectral pansharpening by modeling long-range dependencies. However, their effectiveness is often limited by redundant token representations and a lack of multiscale feature modeling. Hyperspectral images exhibit intrinsic spectral

iFinder: Structured Zero-Shot Vision-Based LLM Grounding for Dash-Cam Video Reasoning

October 1, 2025/https://arxiv.org

Grounding large language models (LLMs) in domain-specific tasks like post-hoc dash-cam driving video analysis is challenging due to their general-purpose training and lack of structured inductive biases. As vision is often the sole modality available for such analysis (i.e., no LiDAR, GPS, etc.), existing

Leveraging Digital Twins for AII-Photonics Networks-as-a-Service: Enabling Innovation and Efficiency

October 1, 2025/ECOC 2025 - The 51st European Conference on Optical Communication, Copenhagen, Denmark

This tutorial presents an architecture and methods for all-photonics networks-as-a-service in distributed Al data center infrastructures. We discuss server-based coherent transceiver architectures, remote transponder control, rapid end-to-end lightpath provisioning, digital longitudinal monitoring,

Computation Stability Tracking Using Data Anchors for Fiber Rayleigh-based Nonlinear Random Projection System

October 1, 2025/ECOC 2025 - The 51st European Conference on Optical Communication

We introduce anchor vectors to monitor Rayleigh-backscattering variability in a fiber-optic computing system that performs nonlinear random projection for image classification. With a ~0.4-s calibration interval, system stability can be maintained with a linear decoder, achieving an average accuracy

Digital Twins Beyond C-band Using GNPy

October 1, 2025/ECOC 2025 - The 51st European Conference on Optical Communication

GNPy advancements enable accurate and efficient modeling of multiband optical networks for digital twin applications. The developed solvers for Kerr nonlinearity and SRS have been validated through simulation and experimentally in C+L transmission, supporting real-world network planning, design, and

End-to-End AI for Distributed Fiber Optics Sensing: Eliminating Intermediate Processing via Raw Data Learning

October 1, 2025/ECOC 2025 - The 51st European Conference on Optical Communication

For the first time, we present an end-to-end AI framework for data analysis in distributed fiber optic sensing. The proposed model eliminates the need for optical phase computation and outperforms traditional data processing pipelines, achieving over 96% recognition accuracy on a diverse acoustic dataset.

Energy-based Generative Models for Distributed Acoustic Sensing Event Classification in Telecom Networks

October 1, 2025/ECOC 2025 - The 51st European Conference on Optical Communication

Distributed fiber-optic sensing combined with machine learning enables continuous monitoring of telecom infrastructure. We employ generative modeling for event classification, supporting semi supervised learning, uncertainty calibration, and noise resilience. Our approach offers a scalable, data-efficient

Observing the Worst- and Best-Case Line-System Transmission Conditions in a C-Band Variable Spectral Load Scenario

October 1, 2025/ECOC 2025 - The 51st European Conference on Optical Communication

We experimentally investigated variable spectral loading in an OMS, identifying performance under best and worst transmission conditions. Metrics and data visualization allowed correlation between channel configurations and OSNR variations, enabling the derivation of a simple spectrum allocation rule.

Optical Network Tomography over Live Production Network in Multi-Domain Environment

October 1, 2025/ECOC 2025 - The 51st European Conference on Optical Communication

We report the first trial of network tomography over a live network in a multi-domain environment. We visualize end-to-end optical powers along multiple routes across multiple domains solely from a commercial B00G transponder, enabling performance bottleneck localization, power and routing optimization,

Utilizing Distributed Acoustic Sensing with Telecom Fibers for Entomological Observations

October 1, 2025/ECOC 2025 - The 51st European Conference on Optical Communication

The 2021 emergence of Brood X cicadas was monitored in situ in our testbed using a DAS system connected to an outdoor telecom fiber over a 16-day period. The spectral and energy characteristics of the cicada calling signal has been measured and analyzed.

200km-Sensing-Range Distributed Acoustic Sensor Link using Enhanced Scattering Fibers

October 1, 2025/ECOC 2025 - The 51st European Conference on Optical Communication

We report a record long 200.6 km distributed acoustic sensing (DAS) link without inline ampli-fication, 28.6% improvement of sensing range has been achieved by using three segments of enhanced-scattering fibre (ESF) with progressively higher scattering enhancements.

Domain Specialization as the Key to Make Large Language Models Disruptive: A Comprehensive Survey

September 3, 2025/ACM Computing Surveys

Large language models (LLMs) have significantly advanced the field of natural language processing (NLP), providing a highly useful, task-agnostic foundation for a wide range of applications. However, directly applying LLMs to solve sophisticated problems in specific domains meets many hurdles, caused

Fiber sensing in IOWN Global Forum

September 1, 2025/Optronics

Fiber sensing function was introduced in 2020 as one of the key technology features for the OpenAPN (all photonics network) developed by IOWN GF (Innovative Optical and Wireless NetworkGlobal Forum) in 2020.To our best knowledge, IOWN GF is the first global standard developmentorganization or technology

Bifröst: Peer-to-peer Load-balancing for Function Execution in Agentic AI Systems

August 25, 2025/31st International European Conference on Parallel and Distributed Computing (EURO-PAR 2025), Dresden, Germany

Agentic AI systems rely on Large Language Models (LLMs) to execute complex tasks by invoking external functions. The efficiency of these systems depends on how well function execution is managed, especially under heterogeneous and high-variance workloads, where function execution times can range from

DiscussLLM: Teaching Large Language Models When to Speak

August 25, 2025/arXiv

Large Language Models (LLMs) have demonstrated remarkable capabilities in understanding and generating human-like text, yet they largely operate as reactive agents, responding only when directly prompted. This passivity creates an “awareness gap,” limiting their potential as truly collaborative partners

Harnessing Vision Models for Time Series Analysis: A Survey

August 18, 2025/The 34th International Joint Conference on Artificial Intelligence (IJCAI 2025 Survey Track)

Time series analysis has witnessed the inspiring development from traditional autoregressive models, deep learning models, to recent Transformers and Large Language Models (LLMs). Efforts in leveraging vision models for time series analysis have also been made along the way but are less visible to the

Multi-modal Time Series Analysis: A Tutorial and Survey

August 7, 2025/31st ACM SIGKDD Conference on Knowledge Discover and Data Mining (ACM KDD 2025)

Multi-modal time series analysis has recently emerged as a prominent research area, driven by the increasing availability of diverse data modalities, such as text, images, and structured tabular data from real-world sources. However, effective analysis of multi-modal time series is hindered by data heterogeneity,

ICeTEA: Mixture of Detectors for Metric-Log Anomaly Detection

August 4, 2025/The 11th Mining and Learning from Time Series Workshop: From Classical Methods to LLMs (KDD MILETS Workshop 2025)

Anomaly detection is essential for identifying unusual system behaviors and has wide-ranging applications, from fraud detection to system monitoring. In web servers, anomalies are typically detected using two types of data: metrics (numerical indicators of performance) and logs (records of system events).

Identifying Combinatorial Regulatory Genes for Cell Fate Decision via Reparameterizable Subset Explanations

August 3, 2025/31st ACM SIGKDD Conference on Knowledge Discover and Data Mining (ACM KDD 2025)

Cell fate decisions are highly coordinated processes governed bycomplex interactions among numerous regulatory genes, whiledisruptions in these mechanisms can lead to developmental abnormalitiesand disease. Traditional methods often fail to capture suchcombinatorial interactions, limiting their ability

Roadside Multi-LiDAR Data Fusion for Enhanced Traffic Safety

August 3, 2025/31st ACM SIGKDD Conference on Knowledge Discover and Data Mining (ACM KDD 2025)

Roadside LiDAR (Light Detection and Ranging) sensors promise safer and faster traffic management and vehicular operations. However, occlusion and small view angles are significant challenges to widespread use of roadside LiDARs. We consider fusing data from multiple LiDARs at a traffic intersection to

Feasibility study on scour monitoring for subsea cables of offshore wind turbines using distributed fiber optic sensors

August 1, 2025/Renewable Energy

Subsea cables are critical components of offshore wind turbines and are subjected to scour. Monitoring the scour conditions of subsea cables plays significant roles in improving safety and operation efficiency and reducing the levelized cost of electricity. This paper presents a feasibility study on

On Synthesizing Data for Context Attribution in Question Answering

August 1, 2025/The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Question Answering (QA) accounts for a significantportion of LLM usage “in the wild”.However, LLMs sometimes produce false ormisleading responses, also known as hallucinations.Therefore, grounding the generatedanswers in contextually provided information—i.e., providing evidence for the generated

Group Relative Augmentation for Data Efficient Action Detection

July 30, 2025/arXiv

Adapting large Video-Language Models (VLMs) for action detection using only a few examples poses challenges like overfitting and the granularity mismatch between scene-level pre-training and required person-centric understanding. We propose an efficient adaptation strategy combining parameter-efficient

Uncertainty Propagation on LLM Agent

July 29, 2025/The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Large language models (LLMs) integrated into multi-step agent systems enable complex decision-making processes across various applications. However, their outputs often lack reliability, making uncertainty estimation crucial. Existing uncertainty estimation methods primarily focus on final-step outputs,

Exploring Multi-Modal Data with Tool-Augmented LLM Agents for Precise Causal Discovery

July 28, 2025/The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Causal discovery is an imperative foundation for decision-making across domains, such as smart health, AI for drug discovery and AIOps. Traditional statistical causal discovery methods, while well-established, predominantly rely on observational data and often overlook the semantic cues inherent in cause-and-effect

EcoDoc: A Cost-Efficient Multimodal Document Processing System for Enterprises Using LLMs

July 27, 2025/The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Enterprises are increasingly adopting Generative AI applications to extract insights from large volumes of multimodal documents in domains such as finance, law, healthcare, and industry. These documents contain structured and unstructured data (images, charts, handwritten texts, etc.) requiring robust

Integration of Fiber Optic Sensing and Sparse Grid Sensors for Accurate Fault Localization in Distribution Systems

July 27, 2025/2025 IEEE PES General Meeting

Fault localization in power distribution networks is essential for rapid recovery and enhancing system resilience. While Phasor Measurement Units (PMUs or ?PMUs) providehigh-resolution measurements for precise fault localization, their widespread deployment is cost-prohibitive. Distributed Fiber Optic

XPF: Agentic AI System for Business Workflow Automation

July 20, 2025/3rd Workshop on AI for Systems (AI4Sys 2025) In conjunction with HPDC 2025

In this paper, we propose a novel agentic AI system called XPF, which enables users to create “agents” using just natural language, where each agent is capable of executing complex, real-world business workflows in an accurate and reliable manner. XPF provides an interface to develop and iterate over

Quantitative Bounds for Length Generalization in Transformers

July 19, 2025/3rd Workshop on High-dimensional Learning Dynamics (HiLD), San Diego, CA

We provide quantitative bounds on the length of sequences required to be observed during training for a transformer to length generalize, e.g., to continue to perform well on sequences unseen during training. Our results improve on Huang et al. [8], who show that there is a finite training length beyond

Re-ranking the Context for Multimodal Retrieval Augmented Generation

July 18, 2025/IR-RAG @ SIGIR25

Retrieval-augmented generation (RAG) enhances large language models (LLMs) by incorporating external knowledge to generate a response within a context with improved accuracy and reduced hallucinations. However, multi-modal RAG systems face unique challenges: (i) the retrieval process may select irrelevant

Beyond the Permutation Symmetry of Transformers: The Role of Rotation for Model Fusion

July 13, 2025/Forty-Second International Conference on Machine Learning (ICML 2025)

Symmetry in the parameter space of deep neural networks (DNNs) has proven beneficial for various deep learning applications. A well-known example is the permutation symmetry in Multi-Layer Perceptrons (MLPs), where permuting the rows of weight matrices in one layer and applying the inverse permutation

PPDiff: Diffusing in Hybrid Sequence-Structure Space for Protein-Protein Complex Design

July 13, 2025/Forty-Second International Conference on Machine Learning (ICML 2025)

Designing protein-binding proteins with high affinity is critical in biomedical research and biotechnology. Despite recent advancements targeting specific proteins, the ability to create high-affinity binders for arbitrary protein targets on demand, without extensive rounds of wet-lab testing,remains

Pathologist-Read vs AI-Driven Assessment of Tumor-Infiltrating Lymphocytes in Melanoma

July 3, 2025/Journal of the American Medical Association Network Open (JAMA Network Open)

Tumor-infiltrating lymphocytes (TILs) are a provocative biomarker in melanoma, influencing diagnosis, prognosis, and immunotherapy outcomes; however, traditional pathologistreadTIL assessment on hematoxylin and eosin–stained slides is prone to interobserver variability, leading to inconsistent clinical

Accelerating Distributed Machine Learning with AllReduce Reconfiguration Based on Optical Circuit Switching

July 1, 2025/OECC/PSC 2025

We propose to apply optical circuit switching to enable dynamic AllReduce reconfiguration for accelerating distributed machine learning. With simulated annealing-based optimization, theproposed AllReduce reconfiguration approach achieves 31% less average training time than existing solutions.

First City-Scale Deployment of DASs with Satellite Imagery and AI for Live Telecom Infrastructure Management

July 1, 2025/OECC/PSC 2025

We demonstrate real-time fiber risk assessment and dynamic network routing in live metro networks using deployed DASs, satellite imagery, and large-scale AI, achieving the first significantreduction in fiber failures in four years

High Definition-Distributed Fiber Optic Sensing and Smart Intersection application

July 1, 2025/OECC/PSC 2025

Distributed fiber optics sensing is applied for traffic management in the intersection. The high-definition fiber sensing data streaming is applied as source and YOLO computer vision model isemployed for event detection classification and localization.

QoT-Driven Control and Optimization in Fiber-Optic WDM Network Systems

July 1, 2025/OECC/PSC 2025

This paper outlines QoT-driven optimization strategies in coherent fiber-optic WDM networks, addressing distinct transmission scenarios, QoT metrics, control-plane methodologies, and emerging trends to enhance network reliability, flexibility and capacity.

Robust Phase Noise Power Spectral Density Estimation Using Multi-Laser Interferometry

July 1, 2025/OECC/PSC 2025

We jointly estimate the phase noise power spectral densities of multiple lasers using interferometry between different combinations of laser pairs. We demonstrate a beat-frequency trackingmethod that allows under-sampling of interferometric products without phase jumps.

Span-based Polarization Sensing in Cables Without Reflectors

July 1, 2025/OECC/PSC 2025

Polarization-based, multi-span sensing over a link without reflection-back circuits is demonstrated experimentally. It is shown that distributed reflection from Rayleigh scattering can serveas an alternative to reflectors after spatial averaging of received state-of-polarization

Toward Intelligent and Efficient Optical Networks: Performance Modeling, Co-existence, and Field Trials

July 1, 2025/OECC/PSC 2025

Optical transmission networks require intelligent traffic adaptation and efficient spectrum usage. We present scalable machine learning (ML) methods for network performance modeling, andfield trials of distributed fiber sensing and classic optical network traffic coexistence.

Distributed Acoustic Sensing Over PON Architecture by Using Enhanced Scattering Fiber

July 1, 2025/Journal of Lightwave Technology

Passive-Optical-Networks (PON) have emerged as a pivotal technology for broadband access network and are now expanding to wireless communication, supporting 5G and development of future 6G frameworks. PON systems are expected to find many new applications, including in electrical power grids, modern

Wavelength tunable distributed vibration sensing over PON architecture using enhanced scattering fiber and ITLA

June 29, 2025/OECC/PSC 2025

We demonstrate a wavelength tunable Distributed-Vibration-Sensing over PON scheme using low-cost ITLA and Enhanced-Scattering-Fibers. Vibrations at frequency grids of 193.40THz and194.60THz in a PON with 1×16 splitter and 21 km feeder-fiber were successfully detected.

Engineered Fibers for Distributed Sensing in Telecom network

June 29, 2025/OECC/PSC 2025

Publication Date: 6/29/2025 Event: OECC/PSC 2025 Reference: WC2-1: 1-3, 2025 Authors: Paul S. Westbrook, OFS Labs; Benyuan Zhu, OFS Labs; Kenneth S. Feder, OFS Labs; Zhou Shi, OFS Labs; Tristan Kremp, OFS Labs; Yaowen Li, NEC Laboratories America, Inc.; Ting Wang, NEC Laboratories America, Inc.; David

Where’s the Liability in the Generative Era? Recovery-based Black-Box Detection of AI-Generated Content

June 14, 2025/The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2025 (CVPR 2025), Nashville, TN

The recent proliferation of photorealistic images created by generative models has sparked both excitement and concern, as these images are increasingly indistinguishable from real ones to the human eye. While offering new creative and commercial possibilities, the potential for misuse, such as in misinformation

SimCache: Similarity Caching for Efficient VLM-based Scene Understanding

June 11, 2025/ELVM Efficient Large Vision Models CVPR Workshop (2nd Edition)

Scene understanding systems analyze visual contexts by detecting objects, their attributes, and the interactions among them to provide a holistic interpretation. Understanding a scene requires analyzing multiple salient regions within a single video frame. Recently, Vision-Language Models (VLMs) have

Solving Inverse Problems via a Score-Based Prior: An Approximation-Free Posterior Sampling Approach

June 5, 2025/arXiv

Diffusion models (DMs) have proven to be effective in modeling high-dimensional distributions, leading to their widespread adoption for representing complex priors in Bayesian inverse problems (BIPs). However, current DM-based posterior sampling methods proposed for solving common BIPs rely on heuristic

Phase-noise Tolerant Per-span Phase and Polarization Sensing

June 2, 2025/SubOptic 2025

Subsea cables include a supervisory system that monitors the health of the amplifier pumps and fiber loss on per span basis. In some of the cables, the monitoring is achieved optically and passively using high-loss loop back paths and wavelength selective reflectors. By sending monitoring pulses through

GFF-Agnostic Black Box Gain Model for non-Flat Input Spectrum

June 2, 2025/SubOptic 2025

We present a simple and accurate semi-analytical model predicting the gain of a single-stage erbium-doped fiber amplifier (EDFA) embedded with an unknown gain flattening filter (GFF). Characteristic wavelength-dependent gain coefficients and their scaling laws are extracted with a limited set of simple

Integration of Distributed Acoustic Sensing and Unrepreatered Transmission for Undersea Cable Monitoring by ESF

June 2, 2025/SubOptic 2025

We present techniques to extend the sensing range in unrepeatered submarine cable systems by utilizing Enhanced-Scattering Fibre (ESF), large-area ultra-low-loss (ULL) fibre, and a digital Distributed Acoustic Sensing (DAS) interrogator. A DAS sensing range of up to 200.6 km has been achieved using 156km

Efficient Semantic Communication Through Transformer-Aided Compression

May 26, 2025/IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN 2025)

Transformers, known for their attention mechanisms, have proven highly effective in focusing on critical elements within complex data. This feature can effectively be used to address the time-varying channels in wireless communication systems. In this work, we introduce a channel-aware adaptive framework

Resilient DFOS Placement Strategy for Power Grid Monitoring: Integrating Fiber and Power Network Dependencies

May 25, 2025/The International Conference on Optical Fiber Sensors OFS-29

We propose a novel Distributed Fiber Optic Sensing (DFOS) placement strategy tailored to the evolving needs of modern power grids, where fiber cables serve dual purposes: communication and real-time sensing. Our approach integrates a heuristic algorithm, PURE (Power Source-aware Route Exploration), with

Detection of Waves and Sea-Surface Vessels via Time Domain Only Analysis of Underwater DAS Data

May 25, 2025/The International Conference on Optical Fiber Sensors OFS-29

A 100-meter-long fiber optic cable was installed at the bottom of a water tank at the Davidson Laboratory, together with a hydrophone for reference. The water tank is approximately 2.5 meters deep and 95 meters long; the tank also employs a 6-paddle wavemaker which can generate programmable surface waves.

Optical Flow Processing for Chirp-Pulse Coherent OTDR

May 25, 2025/The International Conference on Optical Fiber Sensors OFS-29

We propose a novel optical flow processing technique for distributed temperature and strain sensing with the chirped-pulse coherent OTDR. Unlike conventional 1-dimensional cross-correlation methods, the technique treats the 2-dimensional waterfall data as sequential video frames, estimating local shifts

Latency-driven Execution of LLM-generated Application Code on the Computing Continuum

May 19, 2025/The Third Workshop on Urgent Analytics for Distributed Computing (QUICK25) at CCGrid 2025

Latency-critical applications demand quick responses. Ideally, detailed insights are preferable for the best decision making and response actions. However, in situations when detailed insights cannot be provided quickly, even basic information goes a long way in tackling the situation effectively. For

CALIBFREE: Self-Supervised Feature Disentanglement for Calibration-Free Multi-Camera Multi-Object Tracking

May 12, 2025/arXiv

Multi-camera multi-object tracking (MCMOT) faces significant challenges in maintaining consistent object identities across varying camera perspectives, particularly when precise calibration and extensive annotations are required. In this paper, we present CalibFree, a self-supervised representation learning

Multiple Sensor-head Phase-sensitive Optical Time-domain Laser Vibrometer

May 4, 2025/2025 CLEO Conference

We propose a hybrid remote and distributed vibration sensing system based on phase-sensitive optical time-domain reflectometry with collimator-based sensor heads. We demonstrate dual-laser vibrometers that detects nm-scale displacements of remote targets.

Evidence-Based Out-of-Distribution Detection on Multi-Label Graphs

May 3, 2025/SIAM International Conference on Data Mining (SDM 2025), Alexandria, VA

The Out-of-Distribution (OOD) problem in graph-structured data is becoming increasingly important in various areas of research and applications, including social network recommendation [36], protein function detection [9, 21], etc. Furthermore, owing to the inherent multi-label properties of nodes, multi-label

Position Really Matters: Towards a Holistic Approach for Prompt Tuning

April 30, 2025/2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL 2025)

Prompt tuning is highly effective in efficiently extracting knowledge from foundation models, encompassing both language, vision, and vision-language models. However, the efficacy of employing fixed soft prompts with a predetermined position for concatenation with inputs for all instances, irrespective

MixLLM: Dynamic Routing in Mixed Large Language Models

April 29, 2025/2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL 2025)

Large Language Models (LLMs) exhibit potential artificial generic intelligence recently, however, their usage is costly with high response latency. Given mixed LLMs with their own strengths and weaknesses, LLM routing aims to identify the most suitable model for each query in the stream to maximize response

DISC: Dynamic Decomposition Improves LLM Inference Scaling (SSI-FM)

April 28, 2025/ICLR Workshop on Scaling Self-Improving Foundation Models without Human Supervision (SSI-FM) at ICLR 2025

Inference scaling methods often rely on decomposing problems into steps, followed by sampling and selecting the best next steps. However, these steps and their sizes are typically fixed or depend on domain knowledge. We propose dynamic decomposition, a method that adaptively and automatically breaks

DISC: Dynamic Decomposition Improves LLM Inference Scaling (DL4C)

April 28, 2025/Third Workshop on Deep Learning for Code (DL4C) at ICLR 2025

Inference scaling methods often rely on decomposing problems into steps, followed by sampling and selecting the best next steps. However, these steps and their sizes are typically fixed or depend on domain knowledge. We propose dynamic decomposition, a method that adaptively and automatically breaks

F-Fidelity: A Robust Framework for Faithful-NESS Evaluation in Explainable AI

April 28, 2025/The Thirteenth International Conference on Learning Representations

Recent research has developed a number of eXplainable AI (XAI) techniques, such as gradient-based approaches, input perturbation-base methods, and black-box explanation methods. While these XAI techniques can extract meaningful insights from deep learning models, how to properly evaluate them remains

Humanizing the Machine: Proxy Attacks to Mislead LLM Detectors

April 28, 2025/The Thirteenth International Conference on Learning Representations (ICLR 2025)

The advent of large language models (LLMs) has revolutionized the field of text generation, producing outputs that closely mimic human-like writing. Although academic and industrial institutions have developed detectors to prevent the malicious usage of LLM-generated texts, other research has doubt about

SFS: Smarter Code Space Search improves LLM Inference Scaling

April 28, 2025/The Thirteenth International Conference on Learning Representations (ICLR 2025)

We frame code generation as a black-box optimization problem within the code space and demonstrate how optimization-inspired techniques can enhance inference scaling. Based on this perspective, we propose SCATTERED FOREST SEARCH (SFS), a novel approach that improves solution diversity and better exploits

Chain-of-region: Visual Language Models Need Details for Diagram Analysis

April 25, 2025/The Thirteenth International Conference on Learning Representations

Visual Language Models (VLMs) like GPT-4V have broadened the scope of LLM applications, yet they face significant challenges in accurately processing visual details, particularly in scientific diagrams. This paper explores the necessity of meticulous visual detail collection and region decomposition

Progressive Token Length Scaling in Transformer Encoders for Efficient Universal Segmentation

April 24, 2025/ICLR 2025

A powerful architecture for universal segmentation relies on transformers that encode multi-scale image features and decode object queries into mask predictions. With efficiency being a high priority for scaling such models, we observed that the state-of-the-art method Mask2Former uses >50% of its compute

A Quantum Variational Autoencoder Utilizing Regularized Mixed-state Latent Representations

April 11, 2025/Physical Review A

A major challenge in near-term quantum computing is its application to large real-world datasets due to scarce quantum hardware resources. One approach to enabling tractable quantum models for such datasets involves finding low-dimensional representations that preserve essential information for downstream

TSLA: Unified Time Series and Language Model

April 10, 2025/2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)

Real-world time series data often require analysis or interpretation from domain experts. Some tasks, like time series question answering, involve both time series and natural language questions, posing challenges for single-modality language models to understand their interaction. To this end, we present

CLAP-S: Support Set Based Adaptation for Downstream Fiber-optic Acoustic Recognition

April 9, 2025/2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)

Contrastive Language-Audio Pretraining (CLAP) models have demonstrated unprecedented performance in various acoustic signal recognition tasks. Fiber-optic-based acoustic recognition is one of the most important downstream tasks and plays a significant role in environmental sensing. Adapting CLAP for

Text-guided Device-realistic Sound Generation for Fiber-based Sound Event Classification

April 9, 2025/2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)

Recent advancements in unique acoustic sensing devices and large-scale audio recognition models have unlocked new possibilities for environmental sound monitoring and detection. However, applying pretrained models to non-conventional acoustic sensors results in performance degradation due to domain shifts,

Trainingless Adaptation of Pretrained Models for Environmental Sound Classification

April 9, 2025/2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)

Deep neural network (DNN)-based models for environmental sound classification are not robust against a domain to which training data do not belong, that is, out-of-distribution or unseen data. To utilize pretrained models for the unseen domain, adaptation methods, such as finetuning and transfer learning,

LLM-based Distributed Code Generation and Cost-Efficient Execution in the Cloud

April 6, 2025/The Sixteenth International Conference on Cloud Computing, GRIDs, and Virtualization (Cloud Computing 2025)

The advancement of Generative Artificial Intelligence (AI), particularly Large Language Models (LLMs), is reshaping the software industry by automating code generation. Many LLM-driven distributed processing systems rely on serial code generation constrained by predefined libraries, limiting flexibility

1.2 Tb/s/l Real Time Mode Division Multiplexing Free Space Optical Communication with Commercial 400G Open and Disaggregated Transponders

April 3, 2025/OFC 2025

We experimentally demonstrate real time mode division multiplexing free space optical communication with commercial 400G open and disaggregated transponders. As proof of concept,using HG00, HG10, and HG01 modes, we transmit 1.2 Tb/s/l (3´1l´400Gb/s) error free.

DiffOptics: A Conditional Diffusion Model for Fiber Optics Sensing Data Imputation

April 3, 2025/OFC 2025

We present a generative AI framework based on a conditional diffusion model for distributed acoustic sensing (DAS) data imputation. The proposed DiffOptics model generates high-quality DAS data of various acoustic events using telecom fiber cables.

Dual Privacy Protection for Distributed Fiber Sensing with Disaggregated Inference and Fine-tuning of Memory-Augmented Networks

April 3, 2025/OFC 2025

We propose a memory-augmented model architecture with disaggregated computation infrastructure for fiber sensing event recognition. By leveraging geo-distributed computingresources in optical networks, this approach empowers end-users to customize models while ensuring dual privacy protection.

Enhancing EDFAs Greybox Modeling in Optical Multiplex Sections Using Few-Shot Learning

April 3, 2025/OFC 2025

We combine few-shot learning and grey-box modeling for EDFAs in optical lines, training a single EDFA model on 500 spectral loads and transferring it to other EDFAs using 4-8 samples, maintaining low OSNR prediction error.

Field Tests of AI-Driven Road Deformation Detection Leveraging Ambient Noise over Deployed Fiber Networks

April 3, 2025/OFC 2025

This study demonstrates an AI-driven method for detecting road deformations using Distributed Acoustic Sensing (DAS) over existing telecom fiber networks. Utilizingambient traffic noise, it enables real-time, long-term, and scalable monitoring for road safety.

Field Trials of Manhole Localization and Condition Diagnostics by Using Ambient Noise and Temperature Data with AI in a Real-Time Integrated Fiber Sensing System

April 3, 2025/OFC 2025

Field trials of ambient noise-based automated methods for manhole localization and condition diagnostics using a real-time DAS/DTS integrated system were conducted. Crossreferencingmultiple sensing data resulted in a 94.7% detection rate and enhanced anomaly identification.

High-Sensitivity Forward-Transmission Vibration Sensing for Real-World Event Detection in Urban Fiber Networks

April 3, 2025/OFC 2025

Publication Date: 4/3/2025 Event: OFC 2025 Reference: Th4C.2: 1-3, 2025 Authors: Jian Fang, NEC Laboratories America, Inc.; Ming-Fang Huang, NEC Laboratories America, Inc.; Scott Kotrla, Verizon; Tiejun J. Xia, Verizon; Glenn A. Wellbrock, Verizon; Jeffrey A Mundt, Verizon; Ting Wang, NEC Laboratories

Multi-Event Distributed Forwarding Sensing with Dual-Sensor Adaptive Beamforming

April 3, 2025/OFC 2025

We present adaptive beamforming techniques to forward-transmission multi-event vibration sensing in environments with interference and jamming. Experimental validation over 100km fiber demonstrates significant improvements on signal reconstruction, noise reduction, and interference rejection from other

Multi-span OSNR and GSNR Prediction using Cascaded Learning

April 3, 2025/OFC 2025

We implement a cascaded learning framework leveraging three different EDFA and fiber component models for OSNR and GSNR prediction, achieving MAEs of 0.20 and 0.14 dBover a 5-span network under dynamic channel loading.

Optical Line System Physical Digital Model Calibration using a Differential Algorithm

April 3, 2025/OFC 2025

A differential algorithm is proposed to calibrate the physical digital model of an optical line system from scratch at the commissioning phase, using minimal measurements and maximizing signal and OSNR estimation accuracy.

QoT Digital Twin for Bridging Physical Layer Knowledge Gaps in Multi-Domain Networks

April 3, 2025/OFC 2025

We propose building a spectrally resolved QoT Digital Twin for optical network domains where models and telemetry are unavailable, by probing transmission on a singlespectral slot, using GNPy, and demonstrating accurate experimental results.

Scalable Machine Learning Models for Optical Transmission System Management

April 3, 2025/OFC 2025

Optical transmission systems require accurate modeling and performance estimation for autonomous adaption and reconfiguration. We present efficient and scalable machine learning (ML) methods for modeling optical networks at component- and network-level with minimizeddata collection.

Statistical Assessment of System Margin in Metro Networks Impaired by PDL

April 3, 2025/OFC 2025

We experimentally justify the need of analyzing stochastic PDL insertion inboptical metro network nodes. Consequently, we assess conservative OSNR margin comparingdifferent approaches to the case with maxwellian-distributed PDL, through Monte Carlo simulation.

Strain Accumulation Rate in Fiber Spools in the Presence of Ambient Acoustic Noise in Laser Phase Interferometry

April 3, 2025/OFC 2025

We investigate the growth rate of phase power spectral density in fiber spools in the presence of ambient acoustic noise, observing a complex interplay between spool geometry, shielding effects, and phase cancellation at high acoustic frequencies.

Underwater Acoustic OFDM Transmission over Optical Fiber with Distributed Acoustic Sensing

April 3, 2025/OFC 2025

We demonstrate fiber-optic acoustic data transmission using distributed acoustic sensing technology in an underwater environment. An acoustic orthogonal frequencydivisionmultiplexing (OFDM) signal transmitted through a fiber-optic cable deployed in a standard 40-meter-scale underwater testbed.

Variable Temperature and Pump Power Semi-Analytical Gain Model for GFF-Embedded Single-Stage EDFAs

April 3, 2025/OFC 2025

A simple and accurate semi-analytical model for predicting the gain of a single-stage erbium-doped fiber amplifier embedded with an unknown gain flattening filter is proposed for precise system equalization that is crucial for submarine systems.

Real-Time Network-Aware Roadside LiDAR Data Compression

April 2, 2025/Vehicle Technology and Intelligent Transport Systems (VEHITS), 2025

LiDAR technology has emerged as a pivotal tool in Intelligent Transportation Systems (ITS), providing unique capabilities that have significantly transformed roadside traffic applications. However, this transformation comes with a distinct challenge: the immense volume of data generated by LiDAR sensors.

CAMTUNER: Adaptive Video Analytics Pipelines via Real-time Automated Camera Parameter Tuning

March 31, 2025/IEEE Transactions on Mobile Computing Journal

In Video Analytics Pipelines (VAP), Analytics Units (AUs) such as object detection and face recognition operating on remote servers rely heavily on surveillance cameras to capture high-quality video streams to achieve high accuracy. Modern network cameras offer an array of parameters that directly influence

A Smart Sensing Grid for Road Traffic Detection Using Terrestrial Optical Networks and Attention-Enhanced Bi-LSTM

March 31, 2025/Journal of Lightwave Technology

We demonstrate the use of existing terrestrial optical networks as a smart sensing grid, employing a bidirectional long short-term memory (Bi-LSTM) model enhanced with an attention mechanism to detect road vehicles. The main idea of our approach is to deploy a fast, accurate and reliable trained deep

Optimal Single-User Interactive Beam Alignment with Feedback Delay

March 25, 2025/arXiv

Communication in Millimeter wave (mmWave) band relies on narrow beams due to directionality, high path loss, and shadowing. One can use beam alignment (BA) techniques to find and adjust the direction of these narrow beams. In this paper, BA at the base station (BS) is considered, where the BS sends a

400-Gb/s mode division multiplexing-based bidirectional free space optical communication in real-time with commercial transponders

March 21, 2025/Free-Space Laser Communications XXXVII (Conference 13355), Photonics West, SPIE

In this work, for the first time, we experimentally demonstrate mode division multiplexing-based bidirectional free space optical communication in real-time using commercial transponders. As proof of concept, via bidirectional pairs of Hermite-Gaussian modes (HG00, HG10, and HG01), using a Telecom Infra

Free-Space Optical Sensing Using Vector Beam Spectra

March 21, 2025/Complex Light and Optical Forces XIX (Conference 13393), Photonics West, SPIE

Vector beams are spatial modes that have spatially inhomogeneous states of polarization. Any light beam is a linear combination of vector beams, the coefficients of which comprise a vector beam spectrum. In this work, through numerical calculations, a novel method of free-space optical sensing is

EdgeSync: Efficient Edge-Assisted Video Analytics via Network Contention-Aware Scheduling

March 17, 2025/4th IEEE Workshop on Pervasive and Resource-constrained Artificial Intelligence (PeRConAI 2025) - part of IEEE Percom 2025

With the advancement of 5G, edge-assisted video analytics has become increasingly popular, driven by the technologys ability to support low-latency, high-bandwidth applications. However, in scenarios where multiple clients competing for network resources, network contention poses a significant challenge.

Attribute-Centric Compositional Text-to-Image Generation

March 13, 2025/International Journal of Computer Vision

Despite the recent impressive breakthroughs in text-to-image generation, generative models have difficulty in capturing thedata distribution of underrepresented attribute compositions while over-memorizing overrepresented attribute compositions,which raises public concerns about their robustness and

G-Litter Marine Litter Dataset Augmentation with Diffusion Models and Large Language Models on GPU Acceleration

March 12, 2025/Applications, Libraries, and Tools for Computational Science and Machine Learning on Heterogeneous HPC Environments Workshop at PDP 2025

Marine litter detection is crucial for environmental monitoring, yet the imbalance in existing datasets limits model performance in identifying various types of waste accurately. This paper presents an efficient data augmentation pipeline that combines generative diffusion models (e.g., Stable Diffusion)

TimeCAP: Learning to Contextualize, Augment, and Predict Time Series Events with Large Language Model Agents

March 4, 2025/The 39th Annual AAAI Conference on Artificial Intelligence (AAAI 2025)

Time series data is essential in various applications, including climate modeling, healthcare monitoring, and financial analytics. Understanding the contextual information associated with real-world time series data is often essential for accurate and reliable event predictions. In this paper, we introduce

Learning Disentangled Equivariant Representation for Explicitly Controllable 3D Molecule Generation

March 4, 2025/The 39th Annual AAAI Conference on Artificial Intelligence

We consider the conditional generation of 3D drug-like molecules with explicit control over molecular properties such as drug-like properties (e.g., Quantitative Estimate of Druglikenessor Synthetic Accessibility score) and effectively binding to specific protein sites. To tackle this problem, we propose

Exploiting VLM Localizability and Semantics for Open Vocabulary Action Detection (WACV)

March 4, 2025/WACV 2025

Action detection aims to detect (recognize and localize) human actions spatially and temporally in videos. Existing approaches focus on the closed-set setting where an action detector is trained and tested on videos from a fixed set of action categories. However, this constrained setting is not viable

Incident Diagnosing and Reporting System based on Retrieval Augmented Large Language Model

March 3, 2025/The 39th Annual AAAI Conference on Artificial Intelligence (AAAI 2025)

The Internet-of-Things (IoT) is widely used in many applications such as smart city, transportation, healthcare, and environment monitoring. A key task of IoT maintenance is to analyze the abnormal sensor records and generate incident report. Traditionally, domain experts engage in such labor intensive

Domain-Guided Weight Modulation for Semi-Supervised Domain Generalization

March 3, 2025/WACV 2025

Unarguably deep learning models capable of generalizing to unseen domain data while leveraging a few labels are of great practical significance due to low developmental costs. In search of this endeavor we study the challenging problem of semi-supervised domain generalization (SSDG) where the goal is

Reducing Hallucinations of Medical Multimodal Large Language Models with Visual Retrieval-Augmented Generation

February 25, 2025/AAAI 2025, GenAI4Health Workshop

Multimodal Large Language Models (MLLMs) have shown impressive performance in vision and text tasks. However, hallucination remains a major challenge, especially in fields like healthcare where details are critical. In this work, we show how MLLMs may be enhanced to support Visual RAG (V-RAG), a retrieval-augmented

Spatial Signatures for Predicting Immunotherapy Outcomes Using Multi-Omics in Non-Small Cell Lung Cancer

February 1, 2025/Nature Genetics

Non-small cell lung cancer (NSCLC) shows variable responses to immunotherapy, highlighting the need for biomarkers to guide patient selection. We applied a spatial multi-omics approach to 234 advanced NSCLC patients treated with programmed death 1-based immunotherapy across three cohorts to identify

ST-VLM: Kinematic Instruction Tuning for Spatio-Temporal Reasoning in Vision-Language Models

January 17, 2025/arXiv

Spatio-temporal reasoning is essential in understanding real-world environments in various fields, eg, autonomous driving and sports analytics. Recent advances have improved the spatial reasoning ability of Vision-Language Models (VLMs) by introducing large-scale data, but these models still struggle

CLAP-S: Support Set Based Adaptation for Downstream Fiber-optic Acoustic Recognition

January 13, 2025/https://arxiv.org

Contrastive Language-Audio Pretraining (CLAP) models have demonstrated unprecedented performance in various acoustic signal recognition tasks. Fiber optic-based acoustic recognition is one of the most important downstream tasks and plays a significant role in environmental sensing. Adapting CLAP for

RAG-check: Evaluating Multimodal Retrieval Augmented Generation Performance

January 7, 2025/arXiv

Retrieval-augmented generation (RAG) improves large language models (LLMs) by using external knowledge to guide response generation, reducing hallucinations. However, RAG, particularly multi-modal RAG, can introduce new hallucination sources: (i) the retrieval process may select irrelevant pieces (e.g.,

Multi-span optical power spectrum prediction using cascaded learning with one-shot end-to-end measurement

January 1, 2025/Journal of Optical Communications and Networking

Scalable methods for optical transmission performance prediction using machine learning (ML) are studied in metro reconfigurable optical add-drop multiplexer (ROADM) networks. A cascaded learning framework is introduced to encompass the use of cascaded component models for end-to-end (E2E) optical path

Drive-1-to-3: Enriching Diffusion Priors for Novel View Synthesis of Real Vehicles

December 19, 2024/arXiv

The recent advent of large-scale 3D data, e.g. Objaverse, has led to impressive progress in training pose-conditioned diffusion models for novel view synthesis. However, due to the synthetic nature of such 3D data, their performance drops significantly when applied to real-world images. This paper consolidates

Discrete-Continuous Variational Optimization with Local Gradients

December 15, 2024/OPT2024: 16th Annual Workshop on Optimization for Machine Learning (part of NeurIPS 2024)

Variational optimization (VO) offers a general approach for handling objectives which may involve discontinuities, or whose gradients are difficult to calculate. By introducing a variational distribution over the parameter space, such objectives are smoothed, and rendered amenable to VO methods. Local

Subgroup Discovery with the Cox Model

December 15, 2024/NeurIPS 2024 Interpretable AI workshop

We study the problem of subgroup discovery with Cox regression models and introduce a method for finding an interpretable subset of the data on which a Cox model is highly accurate. Our method relies on two technical innovations: the emph (Unknown sysvar: (expected prediction entropy)), a novel metric

VB-LoRA: Extreme Parameter Efficient Fine-Tuning with Vector Banks

December 15, 2024/The Thirty-eighth Annual Conference on Neural Information Processing Systems (NeurIPS 2024)

As the adoption of large language models increases and the need for per-user or per-task model customization grows, the parameter-efficient fine-tuning (PEFT) methods, such as low-rank adaptation (LoRA) and its variants, incur substantial storage and transmission costs. To further reduce stored parameters,

Improving Logits-based Detector without Logits from Black-box LLMs

December 9, 2024/The Thirty-eighth Annual Conference on Neural Information Processing Systems

The advent of Large Language Models (LLMs) has revolutionized text generation, producing outputs that closely mimic human writing. This blurring of lines between machine- and human-written text presents new challenges in distinguishing one from the other a task further complicated by the frequent

Protecting Your LLMs with Information Bottleneck

December 9, 2024/The Thirty-eighth Annual Conference on Neural Information Processing Systems (NeurIPS 2024)

The advent of large language models (LLMs) has revolutionized the field of natural language processing, yet they might be attacked to produce harmful content. Despite efforts to ethically align LLMs, these are often fragile and can be circumvented by jailbreaking attacks through optimized or manual adversarial

DiCE-M: Distributed Code Generation and Execution for Marine Applications – An Edge-Cloud Approach

December 7, 2024/International Workshop on Edge Intelligence in conjunction with ACM SEC 2024

Edge computing has emerged as a transformative technology that reduces application latency, improves cost efficiency, enhances security, and enables large-scale deployment of applications across various domains. In environmental monitoring, systems such as MegaSense[49], use low-cost sensors to gather

Understanding Transcriptional Regulatory Redundancy by Learnable Global Subset Perturbations

December 5, 2024/The 16th Asian Conference on Machine Learning

Transcriptional regulation through cis-regulatory elements (CREs) is crucial for numerous biological functions, with its disruption potentially leading to various diseases. It is well-known that these CREs often exhibit redundancy, allowing them to compensate for each other in response to external

Exploiting VLM Localizability and Semantics for Open Vocabulary Action Detection

December 1, 2024/https://arxiv.org

Action detection aims to detect (recognize and localize) human actions spatially and temporally in videos. Existing approaches focus on the closed-set setting where an action detector is trained and tested on videos from a fixed set of action categories. However, this constrained setting is not viable

Matching Confidences and Softened Target Occurrences for Calibration

November 27, 2024/Digital Image Computing: Techniques & Applications (DICTA 2024)

The problem of calibrating deep neural networks (DNNs) is gaining attention, as these networks are becoming central to many real-world applications. Different attempts have been made to counter the poor calibration of DNNs. Amongst others, train-time calibration methods have unfolded as an effective

A Variational Graph Partitioning Approach to Modeling Protein Liquid-liquid Phase Separation

November 20, 2024/Cell Press (journal family)

Graph neural networks (GNNs) have emerged as powerful tools for representation learning. Their efficacy depends on their having an optimal underlying graph. In many cases, the most relevant information comes from specific subgraphs. In this work, we introduce a GNN-based framework (graph-partitioned

Multi-hop Evidence Pursuit Meets the Web: Team Papelo at FEVER 2024

November 15, 2024/Seventh Fact Extraction and VERification Workshop (FEVER)

Separating disinformation from fact on the web has long challenged both the search and the reasoning powers of humans. We show that the reasoning power of large language models (LLMs) and the retrieval power of modern search engines can be combined to automate this process and explainably verify claims.

Variational methods for Learning Multilevel Genetic Algorithms using the Kantorovich Monad

November 15, 2024/https://arxiv.org

Levels of selection and multilevel evolutionary processes are essential concepts in evolutionary theory, and yet there is a lack of common mathematical models for these core ideas. Here, we propose a unified mathematical framework for formulating and optimizing multilevel evolutionary processes and genetic

A Survey on Detection of LLMs-Generated Content

November 13, 2024/The 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP 2024)

The burgeoning capabilities of advanced large language models (LLMs) such as ChatGPT have led to an increase in synthetic content generation with implications across a variety of sectors, including media, cybersecurity, public discourse, and education. As such, the ability to detect LLMs-generated content

InfuserKI: Enhancing Large Language Models with Knowledge Graphs via Infuser-Guided Knowledge Integration (EMNLP 2024)

November 13, 2024/The 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP 2024)

Large Language Models (LLMs) have achieved exceptional capabilities in open generation across various domains, yet they encounter difficulties with tasks that require intensive knowledge. To address these challenges, methods for integrating knowledge have been developed, which augment LLMs with domain-specific

Large Language Models Can Be Contextual Privacy Protection Learners

November 13, 2024/The 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP 2024)

The proliferation of Large Language Models (LLMs) has driven considerable interest in fine-tuning them with domain-specific data to create specialized language models. Nevertheless, such domain-specific fine-tuning data often contains contextually sensitive personally identifiable information (PII).

Exploring the Role of Reasoning Structures for Constructing Proofs in Multi-Step Natural Language Reasoning with Large Language Models

November 12, 2024/The 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP 2024)

When performing complex multi-step reasoning tasks, the ability of Large Language Models (LLMs) to derive structured intermediate proof steps is important for ensuring that the models truly perform the desired reasoning and for improving models’ explainability. This paper is centered around a focused

Characterization and Modeling of the Noise Figure Ripple in a Dual-Stage EDFA

November 10, 2024/IEEE Photonics Conference (IPC 2024)

The noise figure ripple of a dual-stage EDFA is studied starting from experimental measurements under full spectral load conditions and defining device characteristics. Asemi-analytical model is then proposed showing 0.1 dB standard deviation on the error distribution in all cases of operation.

Enhancing Optical Multiplex Section QoT Estimation Using Scalable Gray-box DNN

November 10, 2024/IEEE Photonics Conference (IPC 2024)

In Optical Multiplex Section (OMS) control and optimization framework, end-to-end (Global) and span-by-span (Local) DNN gray-box strategies are compared in terms of scalability and accuracy of the output signal and noise power predictions. Experimental measurements are carried out in OMSs with increasing

Field Verification of Fault Localization with Integrated Physical-Parameter-Aware Methodology

November 10, 2024/IEEE Photonics Conference (IPC 2024)

We report the first field verification of fault localization in an optical line system (OLS) by integrating digital longitudinal monitoring and OLS calibration, highlighting changes in physical metrics and parameters. Use cases shown are degradation of a fiber span loss and optical amplifier noise figure.

DiCE: Distributed Code generation and Execution

November 5, 2024/The 22nd IEEE International Conference on Pervasive Intelligence and Computing (PICom 2024)

Generative artificial intelligence (GenAI), specifically, Large Language Models (LLMs), have shown tremendous potential in automating several tasks and improving human productivity. Recent works have shown them to be quite useful in writing and summarizing text (articles, blogs, poems, stories, songs,

Transformer-Aided Semantic Communications

October 27, 2024/Asilomar Conference on Signals, Systems, and Computers

The transformer structure employed in large language models (LLMs), as a specialized category of deep neural networks (DNNs) featuring attention mechanisms, stands out for their ability to identify and highlight the most relevant aspects of input data. Such a capability is particularly beneficial in

The WizARd and Apprentice: An Augmented Reality Expert Capture System

October 25, 2024/The 23rd IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2024), Bellevue, WA

Learning to perform physical tasks is ubiquitous yet challenging without expert guidance. While Augmented Reality (AR) has been adopted to overlay instructions directly onto the physical context, the natural authoring of such content remains unexplored. To address this, we developed WizARd and Apprentice,

iRAG: Advancing RAG for Videos with an Incremental Approach

October 21, 2024/The 33rd ACM International Conference on Information and Knowledge Management (CIKM 2024)

Retrieval-augmented generation (RAG) systems combine the strengths of language generation and information retrieval to power many real-world applications like chatbots. Use of RAG for understanding of videos is appealing but there are two critical limitations. One-time, upfront conversion of all content

Optical orbital angular momentum analogy to the Stern-Gerlach experiment

October 1, 2024/Optics Letters

Symmetry breaking has been shown to reveal interesting phenomena in physical systems. A notable example is the fundamental work of Otto Stern and Walther Gerlach [Stern and Zerlach, Z. Physik 9, 349 (1922)] nearly 100 years ago demonstrating a spin angular momentum (SAM) deflection that differed from

Learning to Localize Actions in Instructional Videos with LLM-Based Multi-Pathway Text-Video Alignment

September 29, 2024/The 18th European Conference on Computer Vision ECCV 2024

Learning to localize temporal boundaries of procedure steps in instructional videos is challenging due to the limited availability of annotated large-scale training videos. Recent works focus on learning the cross-modal alignment between video segments and ASR-transcripted narration texts through contrastive

Safe-Sim: Safety-Critical Closed-Loop Traffic Simulation with Diffusion-Controllable Adversaries

September 29, 2024/The 18th European Conference on Computer Vision ECCV 2024

Evaluating the performance of autonomous vehicle planning algorithms necessitates simulating long-tail safety-critical traffic scenarios. However, traditional methods for generating such scenarios often fall short in terms of controllability and realism; they also neglect the dynamics of agent interactions.

TrafficLens: Multi-Camera Traffic Video Analysis Using LLMs

September 24, 2024/27th IEEE International Conference on Intelligent Transportation Systems (ITSC 2024)

Traffic cameras are essential in urban areas, playing a crucial role in intelligent transportation systems. Multiple cameras at intersections enhance law enforcement capabilities, traffic management, and pedestrian safety. However, efficiently managing and analyzing multi-camera feeds poses challenges

Accelerating Distributed Machine Learning with an Efficient AllReduce Routing Strategy

September 23, 2024/Frontiers in Optics 2024, Denver, CO

We propose an efficient routing strategy for AllReduce transfers, which compromise of the dominant traffic in machine learning-centric datacenters, to achieve fast parameter synchronization in distributed machine learning, improving the average training time by 9%.

Extension of the Local-Optimization Global-Optimization (LOGO) Launch Power Strategy to Multi-Band Optical Networks

September 22, 2024/2024 European Conference on Optical Communication (ECOC)

We propose extending the LOGO strategy for launch power settings to multi-band scenarios, maintaining low complexity while addressing key inter-band nonlinear effects and accurate amplifier models. This methodology simplifies multi-band optical multiplex section control, providing an immediate, descriptive

First Field Demonstration of Hollow-Core Fibre Supporting Distributed Acoustic Sensing and DWDM Transmission

September 22, 2024/2024 European Conference on Optical Communication (ECOC)

We demonstrate a method for measuring the backscatter coefficient of hollow-core fibre (HCF), and show the feasibility of distributed acoustic sensing (DAS) with simultaneous 9.6-Tb/s DWDM transmission over a 1.6-km field-deployed HCF cable.

Machine Learning Model for EDFA Predicting SHB Effects

September 22, 2024/2024 European Conference on Optical Communication (ECOC)

Experiments show that machine learning model of an EDFA is capable of modelling spectral hole burning effects accurately. As a result, it significantly outperforms black-box models that neglect inhomogeneous effects. Model achieves a record average RMSE of 0.0165 dB between the model predictions and

Measuring the Transceivers Back-to-Back BER-OSNR Characteristic Using Only a Variable Optical Attenuator

September 22, 2024/2024 European Conference on Optical Communication (ECOC)

We propose a transceiver back-to-back BER-OSNR characterization method that requires only a single VOA; it leverages the receiver SNR degradation caused by received power attenuation. Experiments using commercial transceivers show that the measurement error is less than 0.2 dB in the Q-factor.

Remote Sensing for Power Grid Fuse Tripping Using AI-Based Fiber Sensing with Aerial Telecom Cables

September 22, 2024/2024 European Conference on Optical Communication (ECOC)

For the first time, we demonstrate remote sensing of pole-mounted fuse-cutout blowing in a power grid setup using telecom fiber cable. The proposed frequency-based AI model achieves over 98% detection accuracy using distributed fiber sensing data.

Predicting Spatially Resolved Gene Expression via Tissue Morphology using Adaptive Spatial GNNs (ECCB)

September 16, 2024/2024 European Conference on Computational Biology (ECCB)

Spatial transcriptomics technologies, which generate a spatial map of gene activity, can deepen the understanding of tissue architecture and its molecular underpinnings in health and disease. However, the high cost makes these technologies difficult to use in practice. Histological images co-registered

OPENCAM: Lensless Optical Encryption Camera

September 5, 2024/IEEE Transactions on Computational Imaging

Lensless cameras multiplex the incoming light before it is recorded by the sensor. This ability to multiplex the incoming light has led to the development of ultra-thin, high-speed, and single-shot 3D imagers. Recently, there have been various attempts at demonstrating another useful aspect of lensless

MCTR: Multi Camera Tracking Transformer

September 1, 2024/https://arxiv.org

Multi-camera tracking plays a pivotal role in various real-world applications. While end-to-end methods have gained significant interest in single-camera tracking, multi-camera tracking remains predominantly reliant on heuristic techniques. In response to this gap, this paper introduces Multi-Camera

PAIL: Performance based Adversarial Imitation Learning Engine for Carbon Neutral Optimization

August 29, 2024/30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2024)

Achieving carbon neutrality within industrial operations has become increasingly imperative for sustainable development. It is both a significant challenge and a key opportunity for operational optimization in industry 4.0. In recent years, Deep Reinforcement Learning (DRL) based methods offer promising

InfuserKI: Enhancing Large Language Models with Knowledge Graphs via Infuser-Guided Knowledge Integration (VLDB 2024)

August 28, 2024/International Workshop on LLM+KG: Data Management Opportunities in Unifying Large Language Models+Knowledge Graphs in conjunction with VLDB 2024, Guangzhou, China

Though Large Language Models (LLMs) have shown remarkable open-generation capabilities across diverse domains, they struggle with knowledge-intensive tasks. To alleviate this issue, knowledge integration methods have been proposed to enhance LLMs with domain-specific knowledge graphs using external modules.

Mastering Long-Tail Complexity on Graphs: Characterization, Learning, and Generalization

August 28, 2024/30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2024)

In the context of long-tail classification on graphs, the vast majority of existing work primarily revolves around the development of model debiasing strategies, intending to mitigate class imbalances and enhance the overall performance. Despite the notable success, there is very limited literature that

POND: Multi-Source Time Series Domain Adaptation with Information-Aware Prompt Tuning

August 27, 2024/30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2024)

Time series domain adaptation stands as a pivotal and intricate challenge with diverse applications, including but not limited to human activity recognition, sleep stage classification, and machine fault diagnosis. Despite the numerous domain adaptation techniques proposed to tackle this complex problem,

Semi-Automatic Line-System Provisioning with Integrated Physical-Parameter-Aware Methodology: Field Verification and Operational Feasibility

August 22, 2024/Journal of Optical Communications and Networking

We propose methods and an architecture to conduct measurements and optimize newly installed optical fiber line systems semi-automatically using integrated physics-aware technologies in a data center interconnection (DCI) transmission scenario. We demonstrate, for the first time to our knowledge, digital

Distantly-Supervised Joint Extraction with Noise-Robust Learning

August 16, 2024/The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024), Bangkok, Thailand

Joint entity and relation extraction is a process that identifies entity pairs and their relations using a single model. We focus on the problem of joint extraction in distantly-labeled data,whose labels are generated by aligning entity mentions with the corresponding entity and relation tags using a

Spatially Informed Gene Signatures for Response to Immunotherapy in Melanoma

August 15, 2024/Clinical Cancer Research

We aim to improve the prediction of response or resistance to immunotherapies in patients with melanoma. This goal is based on the hypothesis that current gene signatures predicting immunotherapy outcomes show only modest accuracy due to the lack of spatial information about cellular functions and molecular

Towards Counterfactual Fairness-aware Domain Generalization in Changing Environments

August 9, 2024/IJCAI 2024 - The 33rd International Joint Conference on Artificial Intelligence, Jeju, South Korea

Recognizing domain generalization as a commonplace challenge in machine learning, data distribution might progressively evolve across a continuum of sequential domains in practical scenarios. While current methodologies primarily concentrate on bolstering model effectiveness within these new domains,

First Field Trial of Hybrid Fiber Sensing with Data Transmission Resulting in Enhanced Sensing Sensitivity and Spatial Resolution

August 4, 2024/16th Pacific Rim Conference on Lasers and Electro-Optics (CLEO-PR 2024)

Optical fiber cables, initially designed for telecommunications, are increasingly repurposed for environmental monitoring using distributed fiber sensing technologies [1,2]. Distributed acoustic sensing (DAS) based on phase optical time domain reflectometry (?-OTDR) of Rayleigh backscatter enables various

zeta-QVAE: A Quantum Variational Autoencoder utilizing Regularized Mixed-state Latent Representations

August 1, 2024/https://arxiv.org

A major challenge in near-term quantum computing is its application to large real-world datasets due to scarce quantum hardware resources. One approach to enabling tractable quantum models for such datasets involves compressing the original data to manageable dimensions while still representing essential

DFA-RAG: Conversational Semantic Router for Large Language Model with Definite Finite Automaton

July 27, 2024/The Forty-first International Conference on Machine Learning (ICML 2024), Vienna, Austria

This paper introduces the retrieval-augmented large language model with Definite Finite Automaton (DFA-RAG), a novel framework designed to enhance the capabilities of conversational agents using large language models (LLMs). Traditional LLMs face challenges in generating regulated and compliant responses

RIO-CPD: A Riemannian Geometric Method for Correlation-aware Online Change Point Detection

July 25, 2024/Geometry-grounded Representation Learning and Generative Modeling Workshop (ICML 2024)

The objective of change point detection is to identify abrupt changes at potentially multiple points within a data sequence. This task is particularly challenging in the online setting where various types of changes can occur, including shifts in both the marginal and joint distributions of the data.

Low-Latency Passive Thermal Stabilization of a Silicon Micro-Ring Resonator with Self-Heating

July 25, 2024/APL Photonics

Analog photonic information processing can be implemented with low chip area using wavelength-division multiplexed systems, which typically manipulate light using micro-ring resonators. Micro-rings are uniquely susceptible to thermal crosstalk, with negative system performance consequences if not addressed.

Monitoring AI-Modified Content at Scale: A Case Study on the Impact of ChatGPT on AI Conference Peer Reviews

July 21, 2024/The 41st International Conference on Machine Learning (ICML 2024)

We present an approach for estimating the fraction of text in a large corpus which is likely to be substantially modified or produced by a large language model (LLM). Our maximum likelihood model leverages expert-written and AI-generated reference texts to accurately and efficiently examine real-world

Multi-terminal Germanium Photodetector in a Commercial Silicon Photonics Platform

July 15, 2024/IEEE Summer Topicals & Photonics in Switching and Commuting 2024

We report responsivity measurements of a multiterminal photodetection device in a commercial silicon photonics platform. The ratio of measured responsivities is found to track the relative terminal lengths. This can serve as a highly compact optoelectronic tap/diplexer. More importantly, complex biasing

GNPy Experimental Validation in a C+L Multiband Optical Multiplex Section

July 14, 2024/24th International Conference on Transparent Optical Networks

The GNPy quality-of-transmission estimator has undergone improvements and rigorous experimental validation in a C+L multiband transmission scenario. This includes the incorporation of a disaggregated generalized Gaussian noise model, along with advanced modeling of amplifiers and transceivers. The recently

Optical Amplified Line Self-Healing Using GNPy as a Service by the SDN Control

July 14, 2024/24th International Conference on Transparent Optical Networks

A control architecture for a partially disaggregated optical network is proposed using a GNPy-based digital twin for QoT estimation. The proposed implementation enables soft failure mitigation by autonomously adjusting the amplifier working points.

Knowledge-enhanced Prompt Learning for Open-domain Commonsense Reasoning

July 3, 2024/NEC Technical Journal, Special Issue on Revolutionizing Business Practices with Generative AI

Neural language models for commonsense reasoning often formulate the problem as a QA task and make predictions based on learned representations of language after fine-tuning. However, without providing any fine-tuning data and pre-defined answer candidates, can neural language models still answer commonsense

Optimizing LLM API usage costs with novel query-aware reduction of relevant enterprise data

July 3, 2024/NEC Technical Journal, Special Issue on Revolutionizing Business Practices with Generative AI

Costs of LLM API usage rise rapidly when proprietary enterprise data is used as context for user queries to generate more accurate responses from LLMs. To reduce costs, we propose LeanContext, which generates query-aware, compact and AI model-friendly summaries of relevant enterprise data context. This

Foundational Vision-LLM for AI Linkage and Orchestration

July 3, 2024/NEC Technical Journal, Special Issue on Revolutionizing Business Practices with Generative AI

We propose a vision-LLM framework for automating development and deployment of computer vision solutions for pre-defined or custom-defined tasks. A foundational layer is proposed with a code-LLM AI orchestrator self-trained with reinforcement learning to create Python code based on its understanding

LLMs and MI Bring Innovation to Material Development Platforms

July 3, 2024/NEC Technical Journal, Special Issue on Revolutionizing Business Practices with Generative AI

In this paper, we introduce efforts to apply large language models (LLMs) to the field of material development. NEC is advancing the development of a material development platform. By applying core technologies corresponding to two material development steps, namely investigation activities (Read paper/patent)

Pruning as a Domain-specific LLM Extractor

June 20, 2024/2024 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2024), Mexico City, Mexico

Large Language Models (LLMs) have exhibited remarkable proficiency across a wide array of NLP tasks. However, the escalation in model size also engenders substantial deployment costs. While few efforts have explored model pruning techniques to reduce the size of LLMs, they mainly center on general or

Uncertainty Quantification for In-Context Learning of Large Language Models

June 20, 2024/2024 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2024), Mexico City, Mexico

In-context learning has emerged as a groundbreaking ability of Large Language Models (LLMs) and revolutionized various fields by providing a few task-relevant demonstrations in the prompt. However, trustworthy issues with LLMs response, such as hallucination, have also been actively discussed. Existing

Weakly-Supervised Temporal Action Localization with Multi-Modal Plateau Transformers

June 18, 2024/CVPR 2024 3rd Workshop on Learning with Limited Labelled Data for Image and Video Understanding

Weakly Supervised Temporal Action Localization (WSTAL) aims to jointly localize and classify action segments in untrimmed videos with only video level annotations. To leverage video level annotations most existing methods adopt the multiple instance learning paradigm where frame/snippet level action

Taming Self-Training for Open-Vocabulary Object Detection

June 17, 2024/CVPR2024

Recent studies have shown promising performance in open-vocabulary object detection (OVD) by utilizing pseudo labels (PLs) from pretrained vision and language models (VLMs). However, teacher-student self-training, a powerful and widely used paradigm to leverage PLs, is rarely explored for OVD.

AIDE: An Automatic Data Engine for Object Detection in Autonomous Driving

June 17, 2024/CVPR2024

Autonomous vehicle (AV) systems rely on robust perception models as a cornerstone of safety assurance. However, objects encountered on the road exhibit a long-tailed distribution, with rare or unseen categories posing challenges to a deployed perception model. This necessitates an expensive process of

Self-Training Large Language Models for Improved Visual Program Synthesis With Visual Reinforcement

June 17, 2024/CVPR2024

Visual program synthesis is a promising approach to exploit the reasoning abilities of large language models for compositional computer vision tasks. Previous work has used few-shot prompting with frozen LLMs to synthesize visual programs. Training an LLM to write better visual programs is an attractive

Instantaneous Perception of Moving Objects in 3D

June 17, 2024/CVPR2024

The perception of 3D motion of surrounding traffic participants is crucial for driving safety. While existing works primarily focus on general large motions, we contend that the instantaneous detection and quantification of subtle motions is equally important as they indicate the nuances in driving behavior

LidaRF: Delving into Lidar for Neural Radiance Field on Street Scenes

June 17, 2024/CVPR2024

Photorealistic simulation plays a crucial role in applications such as autonomous driving, where advances in neural radiance fields (NeRFs) may allow better scalability through the automatic creation of digital 3D assets. However, reconstruction quality suffers on street scenes due to largely collinear

A Perspective on Deep Vision Performance with Standard Image and Video Codecs

June 17, 2024/AIS: Vision, Graphics and AI for Streaming Workshop at CVPR 2024

Resource-constrained hardware such as edge devices or cell phones often rely on cloud servers to provide the required computational resources for inference in deep vision models. However transferring image and video data from an edge or mobile device to a cloud server requires coding to deal with network

Deep Video Codec Control for Vision Models

June 17, 2024/AIS: Vision, Graphics and AI for Streaming Workshop at CVPR 2024

Standardized lossy video coding is at the core of almost all real-world video processing pipelines. Rate control is used to enable standard codecs to adapt to different network bandwidth conditions or storage constraints. However standard video codecs (e.g. H.264) and their rate control modules aim to

ViTA: An Efficient Video-to-Text Algorithm using VLM for RAG-based Video Analysis System

June 17, 2024/Multimodal Algorithmic Reasoning (MAR) in conjunction with CVPR 2024

Retrieval-augmented generation (RAG) is used in natural language processing (NLP) to provide query-relevant information in enterprise documents to large language models (LLMs). Such enterprise context enables the LLMs to generate more informed and accurate responses. When enterprise data is primarily

Learning from Synthetic Human Group Activities

June 17, 2024/CVPR 2024

The study of complex human interactions and group activities has become a focal point in human-centric computer vision. However, progress in related tasks is often hindered by the challenges of obtaining large-scale labeled datasets from real-world scenarios. To address the limitation, we introduce M3Act,

Why Not Use Your Textbook? Knowledge-Enhanced Procedure Planning of Instructional Videos

June 17, 2024/CVPR 2024

In this paper we explore the capability of an agent to construct a logical sequence of action steps thereby assembling a strategic procedural plan. This plan is crucial for navigating from an initial visual observation to a target visual outcome as depicted in real-life instructional videos. Existing

Improving the Efficiency-Accuracy Trade-off of DETR-Style Models in Practice

June 17, 2024/The 7th Workshop on Efficient Deep Learning for Computer Vision at CVPR 2024

This report aims to provide a comprehensive view on the inference efficiency of DETR-style detection models. We provide the effect of the basic efficiency techniques and identify the factors that are easily applicable yet effectively improve the efficiency-accuracy trade-off. Specifically, we explore

Seeing the Vibration from Fiber-Optic Cables: Rain Intensity Monitoring using Deep Frequency Filtering

June 17, 2024/2024 CVPR Workshop: 20th Workshop on Perception Beyond the Visible Spectrum

The various sensing technologies such as cameras LiDAR radar and satellites with advanced machine learning models offers a comprehensive approach to environmental perception and understanding. This paper introduces an innovative Distributed Fiber Optic Sensing (DFOS) technology utilizing the existing

Generating Enhanced Negatives for Training Language-Based Object Detectors

June 16, 2024/CVPR2024

The recent progress in language-based open-vocabulary object detection can be largely attributed to finding better ways of leveraging large-scale data with free-form text annotations. Training such models with a discriminative objective function has proven successful, but requires good positive and negative

Deep Learning-based Intrusion Detection and Impulsive Event Classification for Distributed Acoustic Sensing across Telecom Networks

June 15, 2024/IEEE Journal of Lightwave Technology

We introduce two pioneering applications leveraging Distributed Fiber Optic Sensing (DFOS) and Machine Learning (ML) technologies. These innovations offer substantial benefits forfortifying telecom infrastructures and public safety. By harnessing existing telecom cables, our solutions excel in perimeter

Deep Learning-Based Real-Time Quality Control of Standard Video Compression for Live Streaming

June 9, 2024/IEEE International Conference on Communication (ICC 2024)

Ensuring high-quality video content for wireless users has become increasingly vital. Nevertheless, maintaining a consistent level of video quality faces challenges due to the fluctuating encoded bitrate, primarily caused by dynamic video content, especially in live streaming scenarios. Video compression

ECO-LLM: LLM-based Edge Cloud Optimization

June 3, 2024/AI4Sys '24 at HPDC 2024

AI/ML techniques have been used to solve systems problems, but their applicability to customize solutions on-the-fly has been limited. Traditionally, any customization required manually changing the AI/ML model or modifying the code, configuration parameters, application settings, etc. This incurs too

StreamingRAG: Real-time Contextual Retrieval and Generation Framework

June 3, 2024/AI4Sys '24 At HPDC 2024

Extracting real-time insights from multi-modal data streams from various domains such as healthcare, intelligent transportation, and satellite remote sensing remains a challenge. High computational demands and limited knowledge scope restrict the applicability of Multi-Modal Large Language Models (MM-LLMs)

Predicting Spatially Resolved Gene Expression via Tissue Morphology using Adaptive Spatial GNNs

June 3, 2024/https://www.biorxiv.org

Motivation Spatial transcriptomics technologies, which generate a spatial map of gene activity, can deepen the understanding of tissue architecture and its molecular underpinnings in health and disease. However, the high cost makes these technologies difficult to use in practice. Histological images

LeanContext: Cost-efficient Domain-specific Question Answering Using LLMs

June 1, 2024/Natural Language Processing

Question-answering (QA) is a significant application of Large Language Models (LLMs), shaping chatbot capabilities across healthcare, education, and customer service. However, widespread LLM integration presents a challenge for small businesses due to the high expenses of LLM API usage. Costs rise rapidly

Advancing Sustainability in Global Supply Chains through Agent-based Simulation

May 30, 2024/The Eighteenth International Conference on Digital Society (ICDS 2024)

In today’s world, with its complex global supply chains, the difficulties and uncertainties we face offer both challenges and opportunities for making things better, especially in terms of efficiency and sustainability. These challenges grow due to unpredictable events, such as natural disasters, unexpected

Improving Test-Time Adaptation For Histopathology Image Segmentation: Gradient-To-Parameter Ratio Guided Feature Alignment

May 28, 2024/21st IEEE International Symposium on Biomedical Imaging (ISBI 2024)

In the field of histopathology, computer-aided systems face significant challenges due to diverse domain shifts. They include variations in tissue source organ, preparation and scanningprotocols. These domain shifts can significantly impact algorithms performance in histopathology tasks, such as cancer

Impeller: A Path-based Heterogeneous Graph Learning Method for Spatial Transcriptomic Data Imputation

May 28, 2024/Bioinformatics

Recent advances in spatial transcriptomics allow spatially resolved gene expression measurements with cellular or even sub-cellular resolution, directly characterizing the complex spatiotemporal gene expression landscape and cell-to-cell interactions in their native microenvironments. Due to technology

Evaluating Cellularity Estimation Methods: Comparing AI Counting with Pathologists’ Visual Estimates

May 28, 2024/Diagnostics

The development of next-generation sequencing (NGS) has enabled the discovery of cancer-specific driver gene alternations, making precision medicine possible. However, accurategenetic testing requires a sufficient amount of tumor cells in the specimen. The evaluation of tumor content ratio (TCR) from

MULAN: Multi-modal Causal Structure Learning and Root Cause Analysis for Microservice Systems

May 17, 2024/The Web Conference 2024 (WWW 2024)

Effective root cause analysis (RCA) is vital for swiftly restoring services, minimizing losses, and ensuring the smooth operation and management of complex systems. Previous data-driven RCA methods, particularly those employing causal discovery techniques, have primarily focused on constructing dependency

Long-HOT: A Modular Hierarchical Approach for Long-Horizon Object Transport

May 13, 2024/ICRA 24, PACIFICO Yokohama, Japan & CVPR2024 Seattle, WA

We aim to address key challenges in long-horizon embodied exploration and navigation by proposing a long-horizon object transport task called Long-HOT and a novel modular framework for temporally extended navigation. Agents in Long-HOT need to efficiently find and pick up target objects that are scattered

DNA-GPT: Divergent N-Gram Analysis for Training-Free Detection of GPT-Generated Text

May 11, 2024/12th International Conference on Learning Representations (ICLR 2024)

Large language models (LLMs) have notably enhanced the fluency and diversity of machine-generated text. However, this progress also presents a significant challenge in detecting the origin of a given text, and current research on detection methods lags behind the rapid evolution of LLMs. Conventional

Improving Open Information Extraction with Large Language Models: A Study on Demonstration Uncertainty

May 11, 2024/ICLR 2024 Workshop on Reliable and Responsible Foundation Models

Open Information Extraction (OIE) task aims at extracting structured facts from unstructured text, typically in the form of (subject, relation, object) triples. Despite the potential of large language models (LLMs) like ChatGPT as a general task solver, they lag behind state-of-the-art (supervised) methods

Parametric Augmentation for Time Series Contrastive Learning

May 11, 2024/12th International Conference on Learning Representations (ICLR 2024)

Modern techniques like contrastive learning have been effectively used in many areas, including computer vision, natural language processing, and graph-structured data. Creating positive examples that assist the model in learning robust and discriminative representations is a crucial stage in contrastive

Towards Robust Fidelity for Evaluating Explainability of Graph Neural Networks

May 11, 2024/12th International Conference on Learning Representations (ICLR 2024)

Graph Neural Networks (GNNs) are neural models that leverage the dependency structure in graphical data via message passing among the graph nodes. GNNs have emerged as pivotal architectures in analyzing graph-structured data, and their expansive application in sensitive domains requires a comprehensive

Deep Learning-Based Real-Time Rate Control for Live Streaming on Wireless Networks

May 8, 2024/IEEE International Conference on Machine Learning for Communication and Networking (IEEE ICMLCN 2024)

Providing wireless users with high-quality video content has become increasingly important. However, ensuring consistent video quality poses challenges due to variable encodedbitrate caused by dynamic video content and fluctuating channel bitrate caused by wireless fading effects. Suboptimal selection

Strategic Preys Make Acute Predators: Enhancing Camouflaged Object Detectors by Generating Camouflaged Objects

May 7, 2024/ICLR 2024

Camouflaged object detection (COD) is the challenging task of identifying camouflaged objects visually blended into surroundings. Albeit achieving remarkable success, existing COD detectors still struggle to obtain precise results in some challenging cases. To handle this problem, we draw inspiration

CLAP: Cost and Latency-Aware Placement of Microservices on the Computing Continuum

May 6, 2024/2nd International Workshop on Urgent Analytics for the Computing Continuum (QUICK '24 co-located with CCGrid 2024)

For microservices-based real-time stream processing applications, computing at the edge delivers fast responses for low workloads, but as workload increases, the response time starts to slow down due to limited compute capacity. Abundant compute capacity in the cloud delivers fast responses even for

Deep Learning Gain and Tilt Adaptive Digital Twin Modeling of Optical Line Systems for Accurate OSNR Predictions

May 6, 2024/ONDM 2024

We propose a deep learning algorithm trained on varied spectral loads and EDFA working points to generate a digital twin of an optical line system able to optimize line control and to enhance OSNR predictions.

Local and Global Optimization Methods for Optical Line Control Based on Quality of Transmission

May 1, 2024/Journal of Optical Communications and Networking

The ever-increasing demand for data traffic in recent decades has pushed network operators to give importance to the aspect of infrastructure control to facilitate its scalability and maximize its capacity. A generic lightpath (LP) is deployed starting from a traffic request between a given pair of nodes

iRAG: An Incremental Retrieval Augmented Generation System for Videos

April 24, 2024/https://arxiv.org

Retrieval augmented generation (RAG) systems combine the strengths of language generation and information retrieval to power many real-world applications like chatbots. Use of RAG for combined understanding of multimodal data such as text, images and videos is appealing but two critical limitations exist:

Radio-Frequency Linear Analysis and Optimization of Silicon Photonic Neural Networks

April 21, 2024/Advanced Photonics Research

Broadband analog signal processors utilizing silicon photonics have demonstrated a significant impact in numerous application spaces, offering unprecedented bandwidths, dynamic range, and tunability. In the past decade, microwave photonic techniques have been applied to neuromorphic processing, resulting

Efficient Transformer Encoders for Mask2Former-style Models

April 16, 2024/https://arxiv.org

Vision transformer based models bring significant improvements for image segmentation tasks. Although these architectures offer powerful capabilities irrespective of specific segmentation tasks, their use of computational resources can be taxing on deployed devices. One way to overcome this challenge

Low-rank Constrained Multichannel Signal Denoising Considering Channel-dependent Sensitivity Inspired by Self-supervised Learning for Optical Fiber Sensing

April 14, 2024/ICASSP 2024

Optical fiber sensing is a technology wherein audio, vibrations, and temperature are detected using an optical fiber; especially the audio/vibrations-aware sensing is called distributed acoustic sensing (DAS). In DAS, observed data, which is comprised of multichannel data, has suffered from severe noise

Provable Membership Inference Privacy

April 9, 2024/Transactions on Machine Learning Research

In applications involving sensitive data, such as finance and healthcare, the necessity for preserving data privacy can be a significant barrier to machine learning model development.Differential privacy (DP) has emerged as one canonical standard for provable privacy. However, DPs strong theoretical

Link Loss Analysis of Integrated Linear Weight Bank within Silicon Photonic Neural Network

April 7, 2024/SPIE Photonics Europe 2024

Over the last decade, silicon photonic neural networks have demonstrated the possibility of photonic-enabled machine learning at the edge. These systems enable low-latency ultra-wideband classifications, channel estimations, and many other signal characterization tasks within wireless environments. While

4D Optical Link Tomography: First Field Demonstration of Autonomous Transponder Capable of Distance, Time, Frequency, and Polarization-Resolved Monitoring

March 24, 2024/OFC 2024

We report the first field demonstration of 4D link tomography using a commercial transponder, which offers distance, time, frequency, and polarization-resolved monitoring. This scheme enables autonomous transponders that identify locations of multiple QoT degradation causes.

Field Implementation of Fiber Cable Monitoring for Mesh Networks with Optimized Multi-Channel Sensor Placement

March 24, 2024/OFC 2024

We develop a heuristic solution to effectively optimize the placement of multi-channel distributed fiber optic sensors in mesh optical fiber cable networks. The solution has beenimplemented in a field network to provide continuous monitoring.

Inline Fiber Type Identification using In-Service Brillouin Optical Time Domain Analysis

March 24, 2024/OFC 2024

We proposed the use of BOTDA as a monitoring tool to identify fiber types present in deployed hybrid-span fiber cables, to assist in network planning, setting optimal launch powers, and selecting correct modulation formats.

Modeling the Input Power Dependency in Transceiver BER-ONSR for QoT Estimation

March 24, 2024/OFC 2024

We propose a method to estimate the input power dependency of the transceiver BER-OSNR characteristic. Experiments using commercial transceivers show that estimation error in Q-factor is less than 0.2 dB.

Multi-Span Optical Power Spectrum Prediction using ML-based EDFA Models and Cascaded Learning

March 24, 2024/OFC 2024

We implement a cascaded learning framework using component-level EDFA models for optical power spectrum prediction in multi-span networks, achieving a mean absolute error of 0.17 dB across 6 spans and 12 EDFAs with only one-shot measurement.

Optical Line Physical Parameters Calibration in Presence of EDFA Total Power Monitors

March 24, 2024/OFC 2024

A method is proposed in order to improve QoT-E by calibrating the physical model parameters of an optical link post-installation, using only total power monitors integrated into the EDFAs and an OSA at the receiver.

Optical Network Anomaly Detection and Localization Based on Forward Transmission Sensing and Route Optimization

March 24, 2024/OFC 2024

We introduce a novel scheme to detect and localize optical network anomaly using forward transmission sensing, and develop a heuristic algorithm to optimize the route selection. The performance is verified via simulations and network experiments.

Improving Real-time Data Streams Performance on Autonomous Surface Vehicles using DataX

March 20, 2024/The 32nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP 2024)

In the evolving Artificial Intelligence (AI) era, the need for real-time algorithm processing in marine edge environments has become a crucial challenge. Data acquisition, analysis, and processing in complex marine situations require sophisticated and highly efficient platforms. This study optimizes

LARA: Latency-Aware Resource Allocator for Stream Processing Applications

March 20, 2024/The 32nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP 2024)

One of the key metrics of interest for stream processing applications is latency, which indicates the total time it takes for the application to process and generate insights from streaming input data. For mission-critical video analytics applications like surveillance and monitoring, it is of paramount

Dynamic Causal Discovery in Imitation Learning

March 4, 2024/The 17th ACM International Conference on Web Search and Data Mining (WSDM 2024), Merida, Yucatan, Mexico

Imitation learning, which learns agent policy by mimicking expert demonstration, has shown promising results in many applications such as medical treatment regimes and self-driving vehicles. However, it remains a difficult task to interpret control policies learned by the agent. Difficulties mainly come

Self-Consistent Decoding for More Factual Open Responses

February 29, 2024/https://arxiv.org

Self-consistency has emerged as a powerful method for improving the accuracy of short answers generated by large language models. As previously defined, it only concerns the accuracy of a final answer parsed from generated text. In this work, we extend the idea to open response generation, by integrating

Distributed Fiber Optic Sensing for Fault Localization Caused by Fallen Tree Using Physics-informed ResNet

February 19, 2024/2024 ISGT North America

Falling trees or their limbs can cause power lines to break or sag, sometimes resulting in devastating wildfires. Conventional protections such as circuit breakers, overcurrent relays and automatic circuit reclosers may clear short circuits caused by tree contact, but they may not detect cases where

Field Trial of Coexistence and Simultaneous Switching of Real-Time Fiber Sensing and Coherent 400 GbE in a Dense Urban Environment

February 15, 2024/Journal of Lightwave Technology

Recent advances in optical fiber sensing have enabled telecom network operators to monitor their fiber infrastructure while generating new revenue in various application scenarios, including data center interconnect, public safety, smart cities, and seismic monitoring. However, given the high utilization

Fast WDM Provisioning With Minimum Probe Signals: The First Field Experiments For DC Exchanges

February 1, 2024/Journal of Optical Communications and Networking

There are increasing requirements for data center interconnection (DCI) services, which use fiber to connect any DC distributed in a metro area and quickly establish high-capacity optical paths between cloud services and mobile edge computing and the users. In such networks, coherent transceivers with

A system-on-chip microwave photonic processor solves dynamic RF interference in real-time with femtosecond latency

January 15, 2024/Nature Light

Radio-frequency interference is a growing concern as wireless technology advances, with potentially life-threatening consequences like interference between radar altimeters and 5?G cellular networks. Mobile transceivers mix signals with varying ratios over time, posing challenges for conventional digital

Enabling Cooperative Hybrid Beamforming in TDD-based Distributed MIMO Systems

January 6, 2024/IEEE Consumer Communications & Networking Conference (IEEE CCNC 2024)

Distributed massive MIMO networks are envisioned to realize cooperative multi-point transmission in next-generation wireless systems. For efficient cooperative hybrid beamforming, the cluster of access points (APs) needs to obtain precise estimates of the uplink channel to perform reliable downlink precoding.

Differentiable JPEG: The Devil is in The Details

January 3, 2024/IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2024)

JPEG remains one of the most widespread lossy image coding methods. However, the non-differentiable nature of JPEG restricts the application in deep learning pipelines. Several differentiable approximations of JPEG have recently been proposed to address this issue. This paper conducts a comprehensive

Prompt-based Domain Discrimination for Multi-source Time Series Domain Adaptation

December 21, 2023/arxiv.org

Time series domain adaptation stands as a pivotal and intricate challenge with diverse applications, including but not limited to human activity recognition, sleep stage classification, and machine fault diagnosis. Despite the numerous domain adaptation techniques proposed to tackle this complex problem,

Improving Language-Based Object Detection by Explicit Generation of Negative Examples

December 21, 2023/https://arxiv.org

The recent progress in language-based object detection with an open-vocabulary can be largely attributed to finding better ways of leveraging large-scale data with free-form text annotations. Training from image captions with grounded bounding boxes (ground truth or pseudo-labeled) enable the models

Scale Up while Scaling Out Microservices in Video Analytics Pipelines

December 18, 2023/Performance Optimization and Auto-Tuning of Software on Multicore/Manycore Systems (POAT 2023), Singapore

Modern video analytics applications comprise multiple microservices chained together as pipelines and executed on container orchestration platforms like Kubernetes. Kubernetes automatically handles the scaling of these microservices for efficient application execution. There are two popular choices for

Hierarchical Gaussian Mixture based Task Generative Model for Robust Meta-Learning

December 16, 2023/Neural Information Processing Systems (NeurIPS 2023), New Orleans, LA

Meta-learning enables quick adaptation of machine learning models to new tasks with limited data. While tasks could come from varying distributions in reality, most of the existing meta-learning methods consider both training and testing tasks as from the same uni-component distribution, overlooking

Exploring Question Decomposition for Zero-Shot VQA

December 11, 2023/NeurIPS 2023

Visual question answering (VQA) has traditionally been treated as a single-step task where each question receives the same amount of effort, unlike natural human question-answering strategies. We explore a question decomposition strategy for VQA to overcome this limitation. We probe the ability of recently

Open-Ended Commonsense Reasoning with Unrestricted Answer Scope

December 10, 2023/Empirical Methods in Natural Language Processing (EMNLP 2023), Singapore

Open-ended Commonsense Reasoning is defined as solving a commonsense question without providing 1) a short list of answer candidates and 2) a pre-defined answer scope. Conventional ways of formulating the commonsense question into a question-answering form or utilizing external knowledge to learn retrieval-based

Disentangled Wasserstein Autoencoder for T-Cell Receptor Engineering

December 10, 2023/NeurIPS 2023

In protein biophysics, the separation between the functionally important residues (forming the active site or binding surface) and those that create the overall structure (the fold) is a well-established and fundamental concept. Identifying and modifying those functional sites is critical for protein

Weakly-supervised Concealed Object Segmentation with SAM-based Pseudo Labeling and Multi-scale Feature Grouping

December 10, 2023/NeurIPS 2023

Weakly-Supervised Concealed Object Segmentation (WSCOS) aims to segment objects well blended with surrounding environments using sparsely-annotated data for model training. It remains a challenging task since (1) it is hard to distinguish concealed objects from the background due to the intrinsic similarity

DP-Mix: Mixup-based Data Augmentation for Differentially Private Learning

December 10, 2023/NeurIPS 2023

Data augmentation techniques, such as image transformations and combinations, are highly effective at improving the generalization of computer vision models, especially when training data is limited. However, such techniques are fundamentally incompatible with differentially private learning approaches,

Controllable Safety-Critical Closed-Loop Traffic Simulation via Guided Diffusion

December 8, 2023/https://arxiv.org

Evaluating the performance of autonomous vehicle planning algorithms necessitates simulating long-tail traffic scenarios. Traditional methods for generating safety-critical scenarios often fall short in realism and controllability. Furthermore, these techniques generally neglect the dynamics of agent

LLM-ASSIST: Enhancing Closed-Loop Planning with Language-Based Reasoning

December 8, 2023/https://arxiv.org

Although planning is a crucial component of the autonomous driving stack, researchers have yet to develop robust planning algorithms that are capable of safely handling the diverse range of possible driving scenarios. Learning-based planners suffer from overfitting and poor long-tail performance. On

OpEnCam: Optical Encryption Camera

December 8, 2023/https://arxiv.org

Lensless cameras multiplex the incoming light before it is recorded by the sensor. This ability to multiplex the incoming light has led to the development of ultra-thin, high-speed, and single-shot 3D imagers. Recently, there have been various attempts at demonstrating another useful aspect of lensless

Seamless Service Handover in UAV-based Mobile Edge Computing

December 5, 2023/IEEE Global Communications Conference

Unmanned aerial vehicles (UAVs), such as drones, can carry high-performance computing devices (e.g., servers) to provide flexible and on-demand data processing services for theusers in the network edge, leading to the so-called mobile edge computing. In mobile edge computing, researchers have already

Blind Cyclic Prefix-based CFO Estimation in MIMO-OFDM Systems

December 4, 2023/IEEE Global Communications Conference (Globecom 2023)

Low-complexity estimation and correction of carrier frequency offset (CFO) are essential in orthogonal frequency division multiplexing (OFDM). In this paper, we propose a low overhead blind CFO estimation technique based on cyclic prefix (CP), in multi-input multi-output (MIMO)-OFDM systems. We propose

Semantic Multi-Resolution Communications

December 4, 2023/IEEE Globecom 2023 - 3rd Workshop on Semantic Communication for 6G

Deep learning based joint source-channel coding (JSCC) has demonstrated significant advancements in data reconstruction compared to separate source-channel coding (SSCC). This superiority arises from the suboptimality of SSCC when dealing with finite block-length data. Moreover, SSCC falls short in reconstructing

GLAD: Content-Aware Dynamic Graphs for Log Anomaly Detection

December 2, 2023/IEEE International Conference On Knowledge Graph (ICKG-2023), Shanghai, China

Logs play a crucial role in system monitoring and debugging by recording valuable system information, including events and status. Although various methods have been proposed to detect anomalies in log sequences, they often overlook the significance of considering relationships among system components,

Long Term Monitoring and Analysis of Brood X Cicada Activity by Distributed Fiber Optic Sensing Technology

December 1, 2023/Journal of Insect Science

Brood X is the largest of the 15 broods of periodical cicadas, and individuals from this brood emerged across the Eastern United States in spring 2021. Using distributed acoustic sensing (DAS) technology, the activity of Brood X cicadas was monitored in their natural environment in Princeton, NJ. Critical

Real-Time Photonic Blind Interference Cancellation

December 1, 2023/Nature Communications

mmWave devices can broadcast multiple spatially-separated data streams simultaneously in order to increase data transfer rates. Data transfer can, however, be compromised by interference. Photonic blind interference cancellation systems offer a power-efficient means of mitigating interference, but previous

Beyond Communication: Telecom Fiber Networks for Rain Detection and Classification

November 22, 2023/28th International Conference on Optical Fiber Sensors (OFS-28)

We present the field trial of an innovative neural network and DAS-based technique, employing a pre-trained CNN fine-tuning strategy for effective rain detection and classification within two practical scenarios.

Distributed Fiber-Optic Sensor as an Acoustic Communication Receiver Array

November 21, 2023/28th International Conference on Optical Fibers (OFS-28)

A novel acoustic transmission technique using distributed acoustic sensors is introduced. By choosing better incident angles for smaller fading and employing an 8- channel beamformer, over 10KB data is transmitted at a 6.4kbps data rate.

OFDM Signal Transmission Using Distributed Fiber-Optic Acoustic Sensing

November 13, 2023/IEEE Photonics Conference 2023

Acoustic data transmission with the Orthogonal Frequency Division Multiplexing (OFDM) signal has been demonstrated using a Distributed Acoustic Sensor (DAS) based on Phase-sensitive Optical Time-Domain Reflectometry (?-OTDR).

Adaptation Speed Analysis for Fairness-Aware Causal Models

October 25, 2023/32nd ACM International Conference on Information and Knowledge Management (CIKM 2023)

For example, in machine translation tasks, to achieve bidirectional translation between two languages, the source corpus is often used as the target corpus, which involves the training of two models with opposite directions. The question of which one can adapt most quickly to a domain shift is of significant

Calibrate Graph Neural Networks under Out-of-Distribution Nodes via Deep Q-learning

October 25, 2023/32nd ACM International Conference on Information and Knowledge Management (CIKM 2023)

Graph neural networks (GNNs) have achieved great success in dealing with graph-structured data that are prevalent in the real world. The core of graph neural networks is the message passing mechanism that aims to generate the embeddings of nodes by aggregating the neighboring node information. However,

Citizen Science for the Sea with Information Technologies: An Open Platform for Gathering Marine Data and Marine Litter Detection from Leisure Boat Instruments

October 9, 2023/IEEE eScience 2023

Data crowdsourcing is an increasingly pervasive and lifestyle-changing technology due to the flywheel effect that results from the interaction between the Internet of Things and Cloud Computing. This paper presents the Citizen Science for the Sea with Information Technologies (C4Sea-IT) framework. It

Source-Free Domain Adaptive Fundus Image Segmentation with Class-Balanced Mean Teacher

October 9, 2023/MICCAI 2023

This paper studies source-free domain adaptive fundus image segmentation which aims to adapt a pretrained fundus segmentation model to a target domain using unlabeled images. This is a challenging task because it is highly risky to adapt a model only using unlabeled data. Most existing methods tackle

Degradation-Resistant Unfolding Network for Heterogeneous Image Fusion

October 2, 2023/ICCV 2023

Heterogeneous image fusion (HIF) aims to enhance image quality by merging complementary information of images captured by different sensors. Early model-based approaches have strong interpretability while being limited by non-adaptive feature extractors with poor generalizability.

Few-Shot Video Classification via Representation Fusion and Promotion Learning

October 2, 2023/ICCV 2023

Recent few-shot video classification (FSVC) works achieve promising performance by capturing similarity across support and query samples with different temporal alignment strategies or learning discriminative features via Transformer block within each episode. However, they ignore two important issues:

MSI: Maximize Support-Set Information for Few-Shot Segmentation

October 2, 2023/ICCV 2023

Few-Shot Segmentation FSS (Few-shot segmentation) aims to segment a target class using a small number of labeled images (support set). To extract information relevant to the target class, a dominant approach in best performing FSS methods removes background features using a support mask. We observe that

Personalized Semantics Excitation for Federated Image Classification

October 2, 2023/ICCV 2023

Federated learning casts a light on the collaboration of distributed local clients with privacy protected to attain a more generic global model. However, significant distribution shift in input/label space across different clients makes it challenging to well generalize to all clients, which motivates

Domain Generalization Guided by Gradient Signal to Noise Ratio of Parameters

October 2, 2023/ICCV 2023

Overfitting to the source domain is a common issue in gradient-based training of deep neural networks. To compensate for the over-parameterized models, numerous regularization techniques have been introduced such as those based on dropout. While these methods achieve significant improvements on classical

Efficient Controllable Multi-Task Architectures

October 2, 2023/ICCV 2023

We aim to train a multi-task model such that users can adjust the desired compute budget and relative importance of task performances after deployment, without retraining. This enables optimizing performance for dynamically varying user needs, without heavy computational overhead to train and save models

LDP-Feat: Image Features with Local Differential Privacy

October 2, 2023/ICCV 2023

Modern computer vision services often require users to share raw feature descriptors with an untrusted server. This presents an inherent privacy risk, as raw descriptors may be used to recover the source images from which they were extracted. To address this issue, researchers recently proposed privatizing

OmniLabel: A Challenging Benchmark for Language-Based Object Detection

October 2, 2023/ICCV 2023

Language-based object detection is a promising direction towards building a natural interface to describe objects in images that goes far beyond plain category names. While recent methods show great progress in that direction, proper evaluation is lacking. With OmniLabel, we propose a novel task definition,

Long Reach Fibre Optic Distributed Acoustic Sensing using Enhanced Scattering Fibre

October 2, 2023/ECOC 2023

We report significant noise reduction in distributed acoustic sensing (DAS) link using enhanced-scatter fibre (ESF). The longest reach of 195km DAS link without inline amplifications is also demonstrated. We further present demonstration of simultaneous fibre-optic sensing and 400Gb/s data transmissions

First Field Demonstration of Automatic WDM Optical Path Provisioning over Alien Access Links for Data Center Exchange

October 1, 2023/ECOC 2023

We demonstrated under six minutes automatic provisioning of optical paths over field- deployed alien access links and WDM carrier links using commercial-grade ROADMs, whitebox mux-ponders, and multi-vendor transceivers. With channel probing, transfer learning, and Gaussian noise model, we achieved an

Real-time Intrusion Detection and Impulsive Acoustic Event Classification with Fiber Optic Sensing and Deep Learning Technologies over Telecom Networks

October 1, 2023/ECOC 2023

We review various use cases of distributed-fiber-optic-sensing and machine-learning technologies that offer advantages to telecom fiber networks on existing fiber infrastructures. Byleveraging an edge-AI platform, perimeter intrusion detection and impulsive acoustic event classification can be performed

Temporal Graph-Based Incident Analysis System for Internet of Things (ECML)

September 22, 2023/ECML PKDD 2023 - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases

Internet-of-things (IoTs) deploy a massive number of sensors to monitor the system and environment. Anomaly detection on sensor data is an important task for IoT maintenance and operation. In real applications, the occurrence of a system-level incident usually involves hundreds of abnormal sensors, making

Temporal Graph based Incident Analysis System for Internet of Things

September 17, 2023/European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2023)

Internet-of-things (IoTs) deploy a massive number of sensors to monitor the system and environment. Anomaly detection on sensor data is an important task for IoT maintenance and operation. In real applications, the occurrence of a system-level incident usually involves hundreds of abnormal sensors, making

Deep Video Codec Control

August 29, 2023/https://arxiv.org

Deep Video Codec Control Lossy video compression is commonly used when transmitting and storing video data. Unified video codecs (e.g., H.264 or H.265) remain the emph(Unknown sysvar: (de facto)) standard, despite the availability of advanced (neural) compression approaches. Transmitting videos in the

AutoTCL: Automated Time Series Contrastive Learning with Adaptive Augmentations

August 20, 2023/The 32nd International Joint Conference on Artificial Intelligence (IJCAI 2023)

Read AutoTCL: Automated Time Series Contrastive Learning with Adaptive Augmentations publication. Modern techniques like contrastive learning have been effectively used in many areas, including computer vision, natural language processing, and graph-structured data. Creating positive examples that assist

FedSkill: Privacy Preserved Interpretable Skill Learning via Imitation

August 10, 2023/29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023)

Read FedSkill: Privacy Preserved Interpretable Skill Learning via Imitation publication. Imitation learning that replicates experts’ skills via their demonstrations has shown significant success in various decision-making tasks. However, two critical challenges still hinder the deployment of imitation

Incremental Causal Graph Learning for Online Root Cause Localization

August 10, 2023/29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023)

The task of root cause analysis (RCA) is to identify the root causes of system faults/failures by analyzing system monitoring data. Efficient RCA can greatly accelerate system failure recovery and mitigate system damages or financial losses. However, previous research has mostly focused on developing

Interdependent Causal Networks for Root Cause Localization

August 10, 2023/29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

The goal of root cause analysis is to identify the underlying causes of system problems by discovering and analyzing the causal structure from system monitoring data. It is indispensable for maintaining the stability and robustness of large-scale complex systems. Existing methods mainly focus on the

Skill Disentanglement for Imitation Learning from Suboptimal Demonstrations

August 10, 2023/29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2023)

Imitation learning has achieved great success in many sequential decision-making tasks, in which a neural agent is learned by imitating collected human demonstrations. However, existing algorithms typically require a large number of high-quality demonstrations that are difficult and expensive to collect.

State-Aware Anomaly Detection for Massive Sensor Data in Internet of Things

August 7, 2023/The 3rd Workshop on Artificial Intelligence-Enabled Cybersecurity Analytics

With the escalating prevalence of Internet of Things (IoTs) in critical infrastructure, the requirement for efficient and effective anomaly detection solution becomes increasingly important. Unfortunately, most prior research works have largely overlooked to adapt detection criteria for different operational

Automatically Evaluating Opinion Prevalence in Opinion Summarization

August 7, 2023/ECNLP 6 (KDD 2023)

When faced with a large number of product reviews, it is not clear that a human can remember all of them and weight opinions representatively to write a good reference summary. Wepropose an automatic metric to test the prevalence of the opinions that a summary expresses, based on counting the number

Improving Pseudo Labels for Open-Vocabulary Object Detection

August 2, 2023/https://arxiv.org

Recent studies show promising performance in open-vocabulary object detection (OVD) using pseudo labels (PLs) from pretrained vision and language models (VLMs). However, PLs generated by VLMs are extremely noisy due to the gap between the pretraining objective of VLMs and OVD, which blocks further advances

Personalized Federated Learning under Mixture Distributions

July 29, 2023/The 40th International Conference on Machine Learning (ICML 2023)

The recent trend towards Personalized Federated Learning (PFL) has garnered significant attention as it allows for the training of models that are tailored to each client while maintaining data privacy. However, current PFL techniques primarily focus on modeling the conditional distribution heterogeneity

Confidence and Dispersity Speak: Characterizing Prediction Matrix for Unsupervised Accuracy Estimation

July 23, 2023

Confidence and Dispersity Speak: Characterizing Prediction Matrix for Unsupervised Accuracy Estimation This work aims to assess how well a model performs under distribution shifts without using labels. While recent methods study prediction confidence, this work reports prediction dispersity is another

A Temperature-Informed Data-Driven Approach for Behind-the-Meter Solar Disaggregation

July 16, 2023/2023 IEEE Power & Energy Society General Meeting (IEEE PES 2023)

The lack of visibility to behind-the-meter (BTM) PVs causes many challenges to utilities. By constructing a dictionary of typical load patterns based on daily average temperatures and power consumptions, this paper proposes a temperature-informed data-driven approach for disaggregating BTM PV generation.

Retrospective : A Dynamically Configurable Coprocessor For Convolutional Neural Networks

July 1, 2023/ISCA@50 Retrospective: 1996-2020

In 2008, parallel computing posed significant challenges due to the complexities of parallel programming and the bottlenecks associated with efficient parallel execution. Inspired by the remarkable scalability achieved by networking and storage systems in handling extensive packet traffic and persistent

Unsupervised Anomaly Detection Under A Multiple Modeling Strategy Via Model Set Optimization Through Transfer Learning

June 30, 2023/The 26th International Conference on Information Fusion, Charleston, SC

Unsupervised anomaly detection approaches have been widely accepted in applications for industrial systems. Industrial systems often operate with multiple modes since they work for multiple purposes or under different conditions. In order to deal with the difficulty of anomaly detection due to multiple

AnB: Application-In-A-Box To Rapidly Deploy and Self-Optimize 5G Apps

June 26, 2023/International Conference on Smart Computing (SMARTCOMP 2023)

We present Application in a Box (AnB) product concept aimed at simplifying the deployment and operation of remote 5G applications. AnB comes pre-configured with all necessary hardware and software components, including sensors like cameras, hardware and software components for a local 5G wireless network,

Elixir: A System To Enhance Data Quality For Multiple Analytics On A Video Stream

June 26, 2023/International Conference on Smart Computing (IEEE SMARTCOMP 2023)

IoT sensors, especially video cameras, are ubiquitously deployed around the world to perform a variety of computer vision tasks in several verticals including retail, health- care, safety and security, transportation, manufacturing, etc. To amortize their high deployment effort and cost, it is desirable

FactionFormer: Context-Driven Collaborative Vision Transformer Models for Edge Intelligence

June 26, 2023/8th IEEE International Workshop on Smart Service Systems SmartSys 2023 (co-located with SMARTCOMP 2023)

Edge Intelligence has received attention in the recent times for its potential towards improving responsiveness, reducing the cost of data transmission, enhancing security and privacy, and enabling autonomous decisions by edge devices. However, edge devices lack the power and compute resources necessary

Improving Cross-Domain Detection with Self-Supervised Learning

June 19, 2023/CVPR 2023, 2nd Workshop on Learning with Limited Labelled Data for Image and Video Understanding

Cross-Domain Detection (XDD) aims to train a domain-adaptive object detector using unlabeled images from a target domain and labeled images from a source domain. Existing approaches achieve this either by aligning the feature maps or the region proposals from the two domains, or by transferring the style

Camouflaged Object Detection with Feature Decomposition and Edge Reconstruction

June 18, 2023/CVPR 2023

Camouflaged object detection (COD) aims to address the tough issue of identifying camouflaged objects visually blended into the surrounding backgrounds. COD is a challenging task due to the intrinsic similarity of camouflaged objects with the background, as well as their ambiguous boundaries. Existing

Conditional Image-to-Video Generation with Latent Flow Diffusion Models

June 18, 2023/CVPR 2023

Conditional image-to-video (cI2V) generation aims to synthesize a new plausible video starting from an image (e.g., a person’s face) and a condition (e.g., an action class label like smile). The key challenge of the cI2V task lies in the simultaneous generation of realistic spatial appearance and temporal

Exploring Compositional Visual Generation with Latent Classifier Guidance

June 18, 2023/CVPR 2023 - Generative Models for Computer Vision Workshop

Diffusion probabilistic models have achieved enormous success in the field of image generation and manipulation. In this paper, we explore a novel paradigm of using the diffusion model and classifier guidance in the latent semantic space for compositional visual tasks. Specifically, we train latent diffusion

NeurOCS: Neural NOCS Supervision for Monocular 3D Object Localization

June 18, 2023/CVPR 2023

Monocular 3D object localization in driving scenes is a crucial task, but challenging due to its ill-posed nature. Estimating 3D coordinates for each pixel on the object surface holds great potential as it provides dense 2D-3D geometric constraints for the underlying PnP problem. However, high-quality

Q: How to Specialize Large Vision-Language Models to Data-Scarce VQA Tasks? A: Self-Train on Unlabeled Images!

June 18, 2023/CVPR 2023

Finetuning a large vision language model (VLM) on a target dataset after large scale pretraining is a dominant paradigm in visual question answering (VQA). Datasets for specialized tasks such as knowledge-based VQA or VQA in non natural-image domains are orders of magnitude smaller than those for general-purpose

Source-Free Video Domain Adaptation with Spatial-Temporal-Historical Consistency Learning

June 18, 2023/CVPR 2023

Source-free domain adaptation (SFDA) is an emerging research topic that studies how to adapt a pretrained source model using unlabeled target data. It is derived from unsupervised domain adaptation but has the advantage of not requiring labeled source data to learn adaptive models. This makes it particularly

Multi-Label Temporal Evidential Neural Networks for Early Event Detection

June 9, 2023/2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2023)

Early event detection aims to detect events even before the event is complete. However, most of the existing methods focus on an event with a single label but fail to be applied to cases with multiple labels. Another non-negligible issue for early event detection is a prediction with overconfidence due

Beyond One Model Fits All: A Survey of Domain Specialization for Large Language Models

June 9, 2023/arXiv

Large language models (LLMs) have significantly advanced the field of natural language processing (NLP), providing a highly useful, task agnostic foundation for a wide range of applications. The great promise of LLMs as general task solvers motivated people to extend their functionality largely beyond

Towards Realizing the Value of Labeled Target Samples: a Two-Stage Approach for Semi-Supervised Domain Adaptation

June 4, 2023/2023 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2023)

Semi-Supervised Domain Adaptation (SSDA) is a recently emerging research topic that extends from the widely-investigated Unsupervised Domain Adaptation (UDA) by further having a few target samples labeled, i.e., the model is trained with labeled source samples, unlabeled target samples as well as a few

Utility Pole Localization by Learning From Ambient Traces on Distributed Acoustic Sensing

June 4, 2023/2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Utility pole detection and localization is the most fundamental application in aerial-optic cables using distributed acoustic sensing (DAS). The existing pole localization method recognizes the hammer knock signal on DAS traces by learning from knocking vibration patterns. However, it requires many efforts

Real-Time Blind Source Separation with Integrated Photonics for Wireless Signals

May 7, 2023/CLEO 2023

We demonstrate, for the first time, real-time blind source separation of interfering GHz transmitters using photonic weights controlled by an RF-System-on-Chip FPGA. This analog system achieves multi-antenna signal separation with millisecond execution latency.

T-Cell Receptor Optimization with Reinforcement Learning and Mutation Polices for Precision Immunotherapy

April 16, 2023/RECOMB 2023

T cells monitor the health status of cells by identifying foreign peptides displayed on their surface. T-cell receptors (TCRs), which are protein complexes found on the surface of T cells, are able to bind to these peptides. This process is known as TCR recognition and constitutes a key step for immune

StreetAware: A High-Resolution Synchronized Multimodal Urban Scene Dataset

April 3, 2023/Sensors

Access to high-quality data is an important barrier in the digital analysis of urban settings, including applications within computer vision and urban design. Diverse forms of data collected from sensors in areas of high activity in the urban environment, particularly at street intersections, are valuable

Explore Benefits of Distributed Fiber Optic Sensing for Optical Network Service Providers

April 1, 2023/Journal of Lightwave Technology

We review various applications of distributed fiber optic sensing (DFOS) and machine learning (ML) technologies that particularly benefit telecom operators’ fiber networks and businesses. By leveraging relative phase shift of the reflectance of coherent Rayleigh, Brillouin and Raman scattering of light

RIS-aided mmWave Beamforming for Two-way Communications of Multiple Pairs

March 31, 2023/ITU Journal on Future and Evolving Technologies (ITU J-FET), Special issue on Intelligent Suraces and their Applications

Millimeter‑wave (mmWave) communications is a key enabler towards realizing enhanced Mobile Broadband (eMBB) as a key promise of 5G and beyond, due to the abundance of bandwidth available at mmWave bands. An mmWave coverage map consists of blind spots due to shadowing and fading especially in dense

Channel Reciprocity Calibration for Hybrid Beamforming in Distributed MIMO Systems

March 26, 2023/IEEE Wireless Communications and Networking Conference (WCNC 2023), Glasgow, Scotland, UK

Time Division Duplex (TDD)-based distributed massive MIMO systems are envisioned as candidate solution for the physical layer of 6G multi-antenna systems supporting cooperative hybrid beamforming that heavily relies on the obtained uplink channel estimates for efficient coherent downlink precoding. However,

Interpretable Skill Learning for Dynamic Treatment Regimes through Imitation

March 24, 2023/57th Conference on Information Sciences and Systems (CISS 2023)

Imitation learning that mimics experts’ skills from their demonstrations has shown great success in discovering dynamic treatment regimes, i.e., the optimal decision rules to treat an individual patient based on related evolving treatment and covariate history. Existing imitation learning methods,

Data-driven Modelling of EDFAs by Neural Networks

March 13, 2023/SubOptic 2023

Dependence of EDFA gain shape on input power and input spectrum shape is modelled using a simple neural network-based architecture for amplifiers with different gains and output powers. The model can predict the gain within ±0.1 dB. Even though the model has good success predicting the performance of

Improvement of Resilience of Submarine Networks Based on Fiber Sensing

March 13, 2023/SubOptic 2023

Simultaneous phase and polarization sensing with span length resolution using the supervisory path is demonstrated. It is shown that by measuring polarization rotation matrix of the return paths, instead of monitoring only the state of polarization, location of the polarization disturbance can be determined

Dynamic Prompting: A Unified Framework for Prompt Tuning

March 6, 2023/arXiv

It has been demonstrated that prompt tuning is highly effective in efficiently eliciting knowledge from language models (LMs). However, the prompt tuning still lags behind fine tuning, especially when the LMs are small. P tuning v2 (Liu et al., 2021b) makes it comparable with finetuning by adding continuous

Field Trial of Coexistence and Simultaneous Switching of Real-time Fiber Sensing and 400GbE Supporting DCI and 5G Mobile Services

March 5, 2023/OFC 2023

Coexistence of real-time constant-amplitude distributed acoustic sensing (DAS) and 400GbE signals is verified by field trial over metro fibers, demonstrating no QoT impact during co-propagation and supporting preemptive DAS-informed optical path switching before link failure

Polarization Sensing Using Polarization Rotation Matrix Eigenvalue Method

March 5, 2023/OFC 2023

Polarization-based, multi-span sensing over a link with reflection-back circuits is demonstrated experimentally. By measuring rotation matrices instead of just monitoring polarization, a 35 dB extinction in localization is achieved regardless of the disturbance magnitude.

Content-aware auto-scaling of stream processing applications on container orchestration platforms

March 1, 2023/31st Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP 2023)

Modern applications are designed as an interacting set of microservices, and these applications are typically deployed on container orchestration platforms like Kubernetes. Several attractive features in Kubernetes make it a popular choice for deploying applications, and automatic scaling is one such

Exploring the limits of ChatGPT for Query or Aspect based Text Summarization

February 16, 2023/arXiv

Text summarization has been a crucial problem in natural language processing (NLP) for several decades. It aims to condense lengthy documents into shorter versions while retaining the most critical information. Various methods have been proposed for text summarization, including extractive and abstractive

DAS over 1,007-km Hybrid Link with 10-Tb/s DP-16QAM Co-propagation using Frequency-Diverse Chirped Pulses

February 15, 2023/Journal of Lightwave Technology

We report the first distributed acoustic sensing (DAS) experiment with over >1,000 km reach on a hybrid link comprising of a mixture of field and lab fibers with bi-directional inline Raman amplification after each span. We used 20× frequency-diversity chirped-pulses for the probe signal,and recovered

Time Series Contrastive Learning with Information-Aware Augmentations

February 14, 2023/Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI-23)

Various contrastive learning approaches have been proposed in recent years and have achieved significant empirical success. While effective and prevalent, contrastive learning has been less explored for time series data. A key component of contrastive learning is to select appropriate augmentations,

Adversarial Alignment for Source Free Object Detection

February 7, 2023/AAAI 2023

Source-free object detection (SFOD) aims to transfer a detector pre-trained on a label-rich source domain to an unlabeled target domain without seeing source data. While most existing SFOD methods generate pseudo labels via a source-pretrained model to guide training, these pseudo labels usually contain

Ambient Noise based Weakly Supervised Manhole Localization Methods over Deployed Fiber Networks

February 1, 2023/Optics Express

We present a manhole localization method based on distributed fiber optic sensing and weakly supervised machine learning techniques. For the first time to our knowledge, ambient environment data is used for underground cable mapping with the promise of enhancing operational efficiency and reducing field

Drone Detection and Localization using Enhanced Fiber-Optic Acoustic Sensor and Distributed Acoustic Sensing Technology

February 1, 2023/Journal of Lightwave Technology

In recent years, the widespread use of drones has led to serious concerns about safety and privacy. Drone detection using microphone arrays has proven to be a promising method. However, it is challenging for microphones to serve large-scale applications due to the issues of synchronization, complexity,

Distributed fiber optic sensing over readily available telecom fiber networks

January 30, 2023/Photonics West 2023

Distributed Fiber Optic Sensing (DFOS) systems rely on measuring and analyzing different properties of the backscattered light of an optical pulse propagating along a fiber cable. DFOS systems can measure temperature, strain, vibrations, or acoustic excitations on the fiber cable and to their unique

Binding Peptide Generation for MHC Class I Proteins with Deep Reinforcement Learning

January 23, 2023/Bioinformatics

Motivation: MHC Class I protein plays an important role in immunotherapy by presenting immunogenic peptides to anti-tumor immune cells. The repertoires of peptides for various MHC Class I proteins are distinct, which can be reflected by their diverse binding motifs. To characterize binding motifs for

Real-time ConcealedWeapon Detection on 3D Radar Images forWalk-through Screening System

January 3, 2023/WACV 2023

This paper presents a framework for real-time concealed weapon detection (CWD) on 3D radar images for walk-through screening systems. The walk-through screening system aims to ensure security in crowded areas by performing CWD on walking persons, hence it requires an accurate and real-time detection

Split to Learn: Gradient Split for Multi-Task Human Image Analysis

January 3, 2023/WACV23

This paper presents an approach to train a unified deep network that simultaneously solves multiple human-related tasks. A multi-task framework is favorable for sharing information across tasks under restricted computational resources. However, tasks not only share information but may also compete for

On TCR Binding Predictors Failing to Generalize to Unseen Peptides

January 2, 2023/Frontiers in Immunology

Several recent studies investigate TCR-peptide/-pMHC binding prediction using machine learning or deep learning approaches. Many of these methods achieve impressive results on test sets, which include peptide sequences that are also included in the training set. In this work, we investigate how state

DyCo: Dynamic, Contextualized AI Models

December 30, 2022/ACM Transactions on Embedded Computing Systems

Devices with limited computing resources use smaller AI models to achieve low-latency inferencing. However, model accuracy is typically much lower than the accuracy of a bigger model that is trained and deployed in places where the computing resources are relatively abundant. We describe DyCo, a novel

Attentive Variational Information Bottleneck for TCR–peptide interaction prediction

December 26, 2022/Bioinformatics

We present a multi-sequence generalization of Variational Information Bottleneck and call the resulting model Attentive Variational Information Bottleneck (AVIB). Our AVIB model leverages multi-head self-attention to implicitly approximate a posterior distribution over latent encodings conditioned on

Deep Federated Anomaly Detection for Multivariate Time Series Data

December 20, 2022/IEEE BigData 2022 - Special Session 2: Machine Learning on Big Data (MLBD 2022), Osaka, Japan

Although many anomaly detection approaches have been developed for multivariate time series data, limited effort has been made in federated settings in which multivariate time series data are heterogeneously distributed among different edge devices while data sharing is prohibited. In this paper, we

Towards Robust Graph Neural Networks via Adversarial Contrastive Learning

December 20, 2022/2022 IEEE International Conference on Big Data (IEEE BigData 2022), Osaka, Japan

Graph Neural Network (GNN), as a powerful representation learning model on graph data, attracts much attention across various disciplines. However, recent studies show that GNN is vulnerable to adversarial attacks. How to make GNN more robust? What are the key vulnerabilities in GNN? How to address the

KGxBoard: Explainable and Interactive Leaderboard for Evaluation of Knowledge Graph Completion Models

December 7, 2022/2022 Conference on Empirical Methods in Natural Language Processing (EMNLP)

Knowledge Graphs (KGs) store information in the form of (head, predicate, tail)-triples. To augment KGs with new knowledge, researchers proposed models for KG Completion (KGC) tasks such as link prediction, i.e., answering (h, p, ?) or (?, p, t) queries. Such models are usually evaluated with averaged

DeepGAR: Deep Graph Learning for Analogical Reasoning

December 3, 2022/IEEE ICDM 2022 - 22nd IEEE International Conference on Data Mining, Orlando, FL

Analogical reasoning is the process of discovering and mapping correspondences from a target subject to a base subject. As the most well-known computational method of analogical reasoning, Structure-Mapping Theory (SMT) abstracts both target and base subjects into relational graphs and forms the cognitive

Personalized Federated Learning via Heterogeneous Modular Networks

December 3, 2022/IEEE ICDM 2022 - 22nd IEEE International Conference on Data Mining, Orlando, FL

Personalized Federated Learning (PFL) which collaboratively trains a federated model while considering local clients under privacy constraints has attracted much attention. Despite its popularity, it has been observed that existing PFL approaches result in sub-optimal solutions when the joint distribution

Using Global Fiber Networks for Environmental Sensing

November 30, 2022/Proceedings of the IEEE

We review recent advances in distributed fiber optic sensing (DFOS) and their applications. The scattering mechanisms in glass, which are exploited for reflectometry-based DFOS, are Rayleigh, Brillouin, and Raman scatterings. These are sensitive to either strain and/or temperature, allowing optical fiber

APT: Adaptive Perceptual quality based camera Tuning using reinforcement learning

November 29, 2022/The 9th International Conference on Internet of Things: Systems, Management and Security (IOTSMS 2022)

Cameras are increasingly being deployed in cities, enterprises and roads world-wide to enable many applications in public safety, intelligent transportation, retail, healthcare and manufacturing. Often, after initial deployment of the cameras, the environmental conditions and the scenes around these

DataX Allocator: Dynamic resource management for stream analytics at the Edge

November 29, 2022/The 9th International Conference on Internet of Things: Systems, Management and Security (IOTSMS 2022)

Serverless edge computing aims to deploy and manage applications so that developers are unaware of challenges associated with dynamic management, sharing, and maintenance of the edge infrastructure. However, this is a non-trivial task because the resource usage by various edge applications varies based

Availability Analysis for Reliable Distributed Fiber Optic Sensors Placement

November 15, 2022/IEEE Photonics Conference 2022 (IPC 2022)

We perform the availability analysis for various reliable distributed fiber optic sensor placement schemes in the circumstances of multiple failures. The study can help the network carriers to select the optimal protection scheme for their network sensing services, considering both service availability

Distributed Optical Fiber Sensing Using Specialty Optical Fibers

November 13, 2022/2022 IEEE Photonics Conference (IPC)

Distributed fiber optic sensing systems use long section of optical fiber as the sensing media. Therefore, the fiber characteristics determines the sensing capability and performance. In this presentation, various types of specialty optical fibers and their sensing applications will be introduced and

A Multi-sensor Feature Fusion Network Model for Bearings Grease Life Assessment in Accelerated Experiments

November 11, 2022/Neural Computing and Application

This paper presents a multi-sensor feature fusion (MSFF) neural network comprised of two inception layer-type multiple channel feature fusion (MCFF) networks for both inner-sensor and cross-sensor feature fusion in conjunction with a deep residual neural network (ResNet) for accurate grease life assessment

Enhancing Video Analytics Accuracy via Real-time Automated Camera Parameter Tuning

November 7, 2022/The 20th ACM Conference on Embedded Networked Sensor Systems (SenSys 2022)

In Video Analytics Pipelines (VAP), Analytics Units (AUs) such as object detection and face recognition running on remote servers critically rely on surveillance cameras to capture high-quality video streams in order to achieve high accuracy. Modern IP cameras come with a large number of camera parameters

Semi-supervised Identification and Mapping of Water Accumulation Extent using Street-level Monitoring Videos

October 31, 2022/Big Earth Data

Urban flooding is becoming a common and devastating hazard, which causes life loss and economic damage. Monitoring and understanding urban flooding in a highly localized scale is a challenging task due to the complicated urban landscape, intricate hydraulic process, and the lack of high-quality and resolution

The Trade-off between Scanning Beam Penetration and Transmission Beam Gain in mmWave Beam Alignment

October 30, 2022/56th Annual Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA

Beam search algorithms have been proposed to align the beams from an access point to a user equipment. The process relies on sending beams from a set of scanning beams (SB) and tailoring a transmission beam (TB) using the received feedback. In this paper, we discuss a fundamental trade-off between the

Exploiting Unlabeled Data with Vision and Language Models for Object Detection

October 24, 2022/ECCV 2022

Building robust and generic object detection frameworks requires scaling to larger label spaces and bigger training datasets. However, it is prohibitively costly to acquire annotations for thousands of categories at a large scale. We propose a novel method that leverages the rich semantics available

Learning Phase Mask for Privacy-Preserving Passive Depth Estimation

October 24, 2022/ECCV 2022

With over a billion sold each year, cameras are not only becoming ubiquitous, but are driving progress in a wide range of domains such as mixed reality, robotics, and more. However, severe concerns regarding the privacy implications of camera-based solutions currently limit the range of environments

Learning Semantic Segmentation from Multiple Datasets with Label Shifts

October 24, 2022/ECCV 2022

While it is desirable to train segmentation models on an aggregation of multiple datasets, a major challenge is that the label space of each dataset may be in conflict with one another. To tackle this challenge, we propose UniSeg, an effective and model-agnostic approach to automatically train segmentation

Single-Stream Multi-level Alignment for Vision-Language Pretraining

October 24, 2022/ECCV 2022

Self-supervised vision-language pretraining from pure images and text with a contrastive loss is effective, but ignores fine-grained alignment due to a dual-stream architecture that aligns image and text representations only on a global level. Earlier, supervised, non-contrastive methods were capable

Why is the video analytics accuracy fluctuating, and what can we do about it?

October 23, 2022/ECCV 2022 Workshop on Adversarial Robustness in the Real World

It is a common practice to think of a video as a sequence of images (frames), and re-use deep neural network models that are trained only on images for similar analytics tasks on videos. In this paper, we show that this “leap of faith” that deep learning models that work well on images will also

COMPOSER: Compositional Reasoning of Group Activity in Videos with Keypoint-Only Modality

October 17, 2022/ECCV 2022

Group Activity Recognition detects the activity collectively performed by a group of actors, which requires compositional reasoning of actors and objects. We approach the task by modeling the video as tokens that represent the multi-scale semantic concepts in the video. We propose COMPOSER, a Multiscale

Efficient Compression Method for Roadside LiDAR Data

October 17, 2022/31st ACM International Conference on Information and Knowledge Management (CiKM 2022)

Roadside LiDAR (Light Detection and Ranging) sensors are recently being explored for intelligent transportation systems aiming at safer and faster traffic management and vehicular operations. A key challenge in such systems is to efficiently transfer massive point-cloud data from the roadside LiDAR devices

Field Trials of Vibration Detection, Localization and Classification over Deployed Telecom Fiber Cables

October 17, 2022/Frontiers in Optics / Laser Science (FiO LS)

We review sensing fusion results of integrating fiber sensing with video for machine-learning-based localization and classification of impulsive acoustic event detection. Classification accuracy >97% was achieved on aerial coils, and >99% using fiber-based signal enhancers.

Unsupervised Anomaly Detection with Self-Training and Knowledge Distillation

October 16, 2022/IEEE International Conference in Image Processing

Anomaly Detection (AD) aims to find defective patterns or abnormal samples among data, and has been a hot research topic due to various real-world applications. While various AD methods have been proposed, most of them assume the availability of a clean (anomaly-free) training set, which, however, may

5GLoR: 5G LAN Orchestration for Enterprise IoT Applications

October 12, 2022/IEEE Future Networks World Forum 2022

5G-LAN is an enterprise local area network (LAN) that leverages 5G technology for wireless connectivity instead of WiFi. 5G technology is unique: it uses network slicing to distinguish customers in the same traffic class using new QoS technologies in the RF domain. This unique ability is not supported

Analyzing Coreference and Bridging in Product Reviews

October 12, 2022/CRAC 2022, the Fifth Workshop on Computational Models of Reference, Anaphora and Coreference

Product reviews may have complex discourse including coreference and bridging relations to a main product, competing products, and interacting products. Current approaches to aspect-based sentiment analysis (ABSA) and opinion summarization largely ignore this complexity. On the other hand, existing systems

Multi-Faceted Knowledge-Driven Pre-training for Product Representation Learning

September 28, 2022/IEEE Transactions on Knowledge and Data Engineering

As a key component of e-commerce computing, product representation learning (PRL) provides benefits for a variety of applications, including product matching, search, and categorization. The existing PRL approaches have poor language understanding ability due to their inability to capture contextualized

Explainable Anomaly Detection System for Categorical Sensor Data in Internet of Things

September 23, 2022/ECML-PKDD 2022: The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Grenoble, France

Internet of things (IoT) applications deploy massive number of sensors to monitor the system and environment. Anomaly detection on streaming sensor data is an important task for IoT maintenance and operation. However, there are two major challenges for anomaly detection in real IoT applications: (1)

Multi-source Inductive Knowledge Graph Transfer

September 23, 2022/ECML PKDD 2022 - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Grenoble, France

Multi-source Inductive Knowledge Graph Transfer Large-scale information systems, such as knowledge graphs (KGs), enterprise system networks, often exhibit dynamic and complex activities. Recent research has shown that formalizing these information systems as graphs can effectively characterize the entities

Simultaneous Sensing and Communication in Optical Fibers

September 18, 2022/ECOC 2022

We explore two fiber sensing methods which enables coexistence with data transmission on DWDM fiber networks. Vibration detection and localization can be achieved by extracting optical phase from modified coherent transponders. Frequency-diverse chirped-pulse DAS with all-Raman amplification can improve

DataXc: Flexible and efficient communication in microservices-based stream analytics pipelines

September 12, 2022/The 20th IEEE International Conference on Pervasive Intelligence and Computing (PICom 2022)

A big challenge in changing a monolithic application into a performant microservices-based application is the design of efficient mechanisms for microservices to communicate with each other. Prior proposals range from custom point-to-point communication among microservices using protocols like gRPC to

3D Histogram-Based Anomaly Detection for Categorical Sensor Data in Internet of Things

September 9, 2022/VLIoT 2022 - Very Large Internet of Things 2022 (virtual conference)

The applications of Internet-of-things (IoT) deploy a massive number of sensors to monitor the system and environment. Anomaly detection on streaming sensor data is an important task for IoT maintenance and operation. In real IoT applications, many sensors report categorical values rather than numerical

Finite Element Modeling of Pavement and State Awareness Using Fiber Optic Sensing

September 7, 2022/Motion and Vibration Control (MoViC 2022)

A variety of efforts have been put into sensing and modeling of pavements. Such capability is commonly validated with experimental data and used as reference for damage detection and other structural changes. Finite element models (FEM) often provides a high fidelity physics-base benchmark to evaluate

Vibration-Based Status Identification of Power Transmission Poles

September 7, 2022/Motion and Vibration Control (MoViC 2022)

Among the power transmission infrastructures, the low-voltage overhead power lines are specifically critical, due to the complicated roadside environments and the significant number of connections to the end utility users. Maintaining of such a large size grid with mostly wood poles is a challenging

RoVaR: Robust Multi-agent Tracking through Dual-layer Diversity in Visual and RF Sensor Fusion

September 1, 2022/UbiComp 2023 (IMWUT Journal)

The plethora of sensors in our commodity devices provides a rich substrate for sensor-fused tracking. Yet, today’s solutions are unable to deliver robust and high tracking accuracies across multiple agents in practical, everyday environments – a feature central to the future of immersive and collaborative

Application-specific, Dynamic Reservation of 5G Compute and Network Resources by using Reinforcement Learning

August 22, 2022/ACM SIGCOMM 2022 Workshop on Network-Application Integration (NAI 2022)

5G services and applications explicitly reserve compute and network resources in today’s complex and dynamic infrastructure of multi-tiered computing and cellular networking to ensure application-specific service quality metrics, and the infrastructure providers charge the 5G services for the resources

Cosine Similarity based Few-Shot Video Classifier with Attention-based Aggregation

August 22, 2022/26th International Conference on Pattern Recognition (ICPR 2022)

Meta learning algorithms for few-shot video recognition use complex, episodic training but they often fail to learn effective feature representations. In contrast, we propose a new and simpler few-shot video recognition method that does not use meta-learning, but its performance compares well with the

A Deep Learning Framework for Detecting and Localizing Abnormal Pedestrian Behaviors at Grade Crossings

August 18, 2022/Neural Computing and Applications

This paper presents a deep learning-based framework to detect and localize the pedestrians’ anomaly behaviors in videos captured at the grade crossing. A skeleton detection and tracking algorithm are employed to capture the key point trajectories of body movements of the pedestrians. A deep recurrent

CAT: Beyond Efficient Transformer for Content-Aware Anomaly Detection in Event Sequences

August 18, 2022/28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

It is critical and important to detect anomalies in event sequences, which becomes widely available in many application domains. Indeed, various efforts have been made to capture abnormal patterns from event sequences through sequential pattern analysis or event representation learning. However, existing

Towards Learning Disentangled Representations for Time Series

August 18, 2022/28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2022)

Promising progress has been made toward learning efficient time series representations in recent years, but the learned representations often lack interpretability and do not encode semantic meanings by the complex interactions of many latent factors. Learning representations that disentangle these latent

T-Cell Receptor-Peptide Interaction Prediction with Physical Model Augmented Pseudo-Labeling

August 14, 2022/KDD 2022

Predicting the interactions between T-cell receptors (TCRs) and peptides is crucial for the development of personalized medicine and targeted vaccine in immunotherapy. Current datasets for training deep learning models of this purpose remain constrained without diverse TCRs and peptides. To combat the

Rain Intensity Detection and Classification with Pre-existing Telecom Fiber Cables

July 11, 2022/Optical Sensors and Sensing Congress

For the first time, we demonstrate detection and classification of rain intensity using Distributed Acoustic Sensing (DAS). An artificial neural network was applied for rain intensity classification and high precision of over 96% was achieved.

Evolution of Fiber Infrastructure – From Data Transmission to Network Sensing

July 3, 2022/OECC/PSC 2022

We review multiple use cases over deployed networks including co-existing sensing/data transmission, cable cut prevention and perimeter intrusion detection to realize telecom infrastructure can be sensing backbones instead of the sole function of data transmission.

Field Tests of Impulsive Acoustic Event Detection, Localization, and Classification Over Telecom Fiber Networks

July 3, 2022/OECC 2022 PDP

We report distributed-fiber-optic-sensing results on impulsive acoustic events localization/classification over telecom networks. A deep-learning-based model was trained to classify starter-gun and fireworks signatures with high accuracy of > 99% using fiber-based-signal-enhancer and >97% using aerial

Simultaneous Fiber Sensing and Communications

July 3, 2022/OECC/PSC 2022

We review recent advances aimed at increasing the reach of distributed fiber optic sensing with simultaneous data transmission. We review two methods based on measurement of accumulated phase on telecom signals, and chirp-pulsed DAS with inline amplification and frequency diversity.

Template Matching Method with Distributed Acoustic Sensing Data and Simulation Data

July 3, 2022/OECC/PSC 2022

We propose a new method to detect acoustic signals by matching distributed acoustic sensing data with simulation. In the simulation of the dynamic strain on an optical fiber, the optical fiber layouts and the gauge length are properly incorporated. We apply the proposed method to the acoustic-source

Mosaic: Leveraging Diverse Reflector Geometries for Omnidirectional Around-Corner Automotive Radar

July 1, 2022/The 20th ACM International Conference on Mobile Systems, Applications, and Services (MobiSys 2022)

A large number of traffic collisions occur as a result of obstructed sight lines, such that even an advanced driver assistance system would be unable to prevent the crash. Recent work has proposed the use of around-the-corner radar systems to detect vehicles, pedestrians, and other road users in these

StyleT2I: Towards Compositional and High-Fidelity Text-to-Image Synthesis

June 21, 2022/CVPR 2022

Although progress has been made for text-to-image synthesis, previous methods fall short of generalizing to unseen or underrepresented attribute compositions in the input text. Lacking compositionality could have severe implications for robustness and fairness, e.g., inability to synthesize the face

Chimera: Context-Aware Splittable Deep Multitasking Models for Edge Intelligence

June 20, 2022/SMARTCOMP 2022

Design of multitasking deep learning models has mostly focused on improving the accuracy of the constituent tasks, but the challenges of efficiently deploying such models in a device-edge collaborative setup (that is common in 5G deployments) has not been investigated. Towards this end, in this paper,

Controllable Dynamic Multi-Task Architectures

June 19, 2022/CVPR'22

Multi-task learning commonly encounters competition for resources among tasks, specifically when model capacity is limited. This challenge motivates models which allow control over the relative importance of tasks and total compute cost during inference time. In this work, we propose such a controllable

Learning to Learn across Diverse Data Biases in Deep Face Recognition

June 19, 2022/CVPR’22

Convolutional Neural Networks have achieved remarkable success in face recognition, in part due to the abundant availability of data. However, the data used for training CNNs is often imbalanced. Prior works largely focus on the long-tailed nature of face datasets in data volume per identity or focus

MM-TTA: Multi-Modal Test-Time Adaptation for 3D Semantic Segmentation

June 19, 2022/CVPR'22

Test-time adaptation approaches have recently emerged as a practical solution for handling domain shift without access to the source domain data. In this paper, we propose and explore a new multi-modal extension of test-time adaptation for 3D semantic segmentation. We find that, directly applying existing

On Generalizing Beyond Domains in Cross-Domain Continual Learning

June 19, 2022/CVPR'22

Humans have the ability to accumulate knowledge of new tasks in varying conditions, but deep neural networks of-ten suffer from catastrophic forgetting of previously learned knowledge after learning a new task. Many recent methods focus on preventing catastrophic forgetting under the assumption of train

Self-supervised Video Representation Learning with Cascade Positive Retrieval

June 19, 2022/CVPR: Workshop on Learning with Limited Labelled Data for Image and Video Understanding

Self-supervised video representation learning has been shown to effectively improve downstream tasks such as video retrieval and action recognition. In this paper, we present the Cascade Positive Retrieval (CPR) that successively mines positive examples w.r.t. the query for contrastive learning in a

Weakly But Deeply Supervised Occlusion-Reasoned Parametric Road Layouts

June 19, 2022/CVPR'22

We propose an end-to-end network that takes a single perspective RGB image of a complex road scene as input, to produce occlusion-reasoned layouts in perspective space as well as a parametric bird’s-eye-view (BEV) space. In contrast to prior works that require dense supervision such as semantic labels

Distributed Fiber Optic Sensors Placement for Infrastructure-as-a-Sensor

June 17, 2022/Photonic Network Communications

Recently, the distributed fiber optic sensing (DFOS) techniques have advanced rapidly. There emerges various types of DFOS sensors that can monitor physical parameters such as temperature, strain, and vibration. With these DFOS sensors deployed, the telecom networks are capable of offering additional

SEED: Sound Event Early Detection via Evidential Uncertainty

May 27, 2022/2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2022), Singapore (virtual paper presentations)

Sound Event Early Detection (SEED) is an essential task in recognizing the acoustic environments and soundscapes. However, most of the existing methods focus on the offline sound event detection, which suffers from the over-confidence issue of early-stage event detection and usually yield unreliable

Fast Few-shot Debugging for NLU Test Suites

May 26, 2022/Deep Learning Inside Out workshop at ACL 2022

We study few-shot debugging of transformer based natural language understanding models, using recently popularized test suites to not just diagnose but correct a problem. Given a few debugging examples of a certain phenomenon, and a held-out test set of the same phenomenon, we aim to maximize accuracy

Codebook Design for Hybrid Beamforming in 5G Systems

May 16, 2022/IEEE International Conference on Communications (ICC 2022)

Massive MIMO and hybrid beamforming are among the key physical layer technologies for the next generation wireless systems. In the last stage of the hybrid beamforming, the goal is to generate sharp beam with maximal and preferably uniform gain. We highlight the shortcomings of uniform linear arrays

Time Series Prediction and Classification using Silicon Photonic Neuron with Self-Connection

May 15, 2022/CLEO 2022

We experimentally demonstrated the real-time operation of a photonic neuron with a self-connection, a prerequisite for integrated recurrent neural networks (RNNs). After studying two applications, we propose a photonics-assisted platform for time series prediction and classification.

Superclass-Conditional Gaussian Mixture Model for Coarse-To-Fine Few-Shot Learning

April 29, 2022/10th International Conference on Learning Representations (ICLR 2022)

Learning fine-grained embeddings is essential for extending the generalizability of models pre-trained on “coarse” labels (e.g., animals). It is crucial to fields for which fine-grained labeling (e.g., breeds of animals) is expensive, but fine-grained prediction is desirable, such as medicine. The dilemma

ROMA: Resource Orchestration for Microservices-based 5G Applications

April 25, 2022/IEEE/IFIP Network Operations and Management Symposium (NOMS 2022)

With the growth of 5G, Internet of Things (IoT), edge computing and cloud computing technologies, the infrastructure (compute and network) available to emerging applications (AR/VR, autonomous driving, industry 4.0, etc.) has become quite complex. There are multiple tiers of computing (IoT devices, near

Learning Transferable Reward for Query Object Localization with Policy Adaptation

April 25, 2022/Tenth International Conference on Learning Representations (ICLR 2022)

We propose a reinforcement learning-based approach to query object localization, for which an agent is trained to localize objects of interest specified by a small exemplary set. We learn a transferable reward signal formulated using the exemplary set by ordinal metric learning. Our proposed method enables

Provable Adaptation Across Multiway Domains via Representation Learning

April 25, 2022/Tenth International Conference on Learning Representations (ICLR 2022)

This paper studies zero-shot domain adaptation where each domain is indexed on a multi-dimensional array, and we only have data from a small subset of domains. Our goal is to produce predictors that perform well on unseen domains. We propose a model which consists of a domain-invariant latent representation

Opportunistic Temporal Fair Mode Selection and User Scheduling in Full-Duplex Systems

April 16, 2022/JSAC: IEEE Communications Society: Journal of Selected Areas in Communications - Special Issue on Next Generation Multiple Access

In-band full-duplex (FD) communication has emerged as one of the promising techniques to improve data rates in next generation wireless systems. Typical FD scenarios considered in the literature assume FD base stations (BSs) and half-duplex (HD) users activated either in uplink (UL) or downlink (DL),

Codebook Design for Composite Beamforming in Next-generation mmWave Systems

April 10, 2022/IEEE Wireless Communications and Networking Conference (WCNC 2022)

In pursuance of the unused spectrum in higher frequencies, millimeter wave (mmWave) bands have a pivotal role. However, the high path-loss and poor scattering associated with mmWave communications highlight the necessity of employing effective beamforming techniques. In order to efficiently search for

DataXe: A System for Application Self-optimization in Serverless Edge Computing Environments

March 21, 2022/First Workshop on Serverless Computing for Pervasive Cloud-Edge-Device Systems and Services (STARLESS ‘22)

A key barrier to building performant, remotely managed and self-optimizing multi-sensor, distributed stream processing edge applications is high programming complexity. We recently proposed DataX [1], a novel platform that improves programmer productivity by enabling easy exchange, transformations, and

Multi-user Beam Alignment in Presence of Multi-path

March 9, 2022/56th Annual Conference on Information Sciences and Systems (CISS 2022)

To overcome the high pathloss and the intense shadowing in millimeterwave (mmWave) communications, effective beamforming schemes are required which incorporate narrow beams with high beamforming gains. The mm Wave channel consists of a few spatial clusters each associated with an angle of departure (AoD).

DAS over 1,007-km Hybrid Link with 10-Tb/s DP-16QAM Co-propagation using Frequency-Diverse Chirped Pulses (OFC)

March 6, 2022/OFC 2022

We report the first distributed acoustic sensing (DAS) results over>1,000 km on a field-lab hybrid link using chirped-pulses with correlation detection and 20× frequency-diversity, achieving a sensitivity of 100 pa/√Hz at 20-meters spatial resolution.

Distributed Acoustic Sensing for Datacenter Optical Interconnects using Self-Homodyne Coherent Detection

March 6, 2022/OFC 2022

We demonstrate distributed acoustic sensing (DAS) over a bidirectional datacenter link which uses self-homodyne coherent detection for the data signal. Frequency multiplexing allows sharing the optoelectronic hardware, and enables DAS as an auxiliary function.

Employing Fiber Sensing and On-Premise AI Solutions for Cable Safety Protection over Telecom Infrastructure

March 6, 2022/OFC 2022

We review the distributed-fiber-sensing field trial results over deployed telecom networks. With local AI processing, real-time detection, and localization of abnormal events with cable damage threat assessment are realized for cable self-protection.

Perimeter Intrusion Detection with Rayleigh Enhanced Fiber Using Telecom Cables as Sensing Backhaul

March 6, 2022/OFC 2022

We report field test results of facility perimeter intrusion detection with distributed-fiber-sensing technology and backscattering-enhanced-fiber by using deployed telecom fiber cables as sensing backhaul. Various intrusive activities, such as walking/jumping at >100ft distance, are detected.

Remote Drone Detection and Localization with Optical Fiber Microphones and Distributed Acoustic Sensing

March 6, 2022/OFC 2022

We demonstrate the first fiber-optic drone detection method with ultra-highly sensitive optical microphones and distributed acoustic sensor. Accurate drone localization has been achieved through acoustic field mapping and data fusion.

Detection of Road Anomaly Using Distributed Fiber Optic Sensing

March 1, 2022/IEEE Intelligent Transportation Systems Transactions

Road surface condition can significantly impact the interaction between vehicles and pavement structure, which may even cause high fuel consumption and safety issues of drivers and vehicles. Distributed fiber optic sensing (DFOS) technology is a useful tool to perform continuous and real-time monitoring

Vibration Detection and Localization using Modified Digital Coherent Telecom Transponders

March 1, 2022/Journal of Lightwave Technology

We demonstrate a vibration detection and localization scheme based on bidirectional transmission of telecom signals with digital coherent detection at the receivers. Optical phase is extracted from the digital signal processing blocks of the coherent receiver, from which the vibration component is extracted

Weight Pruning Techniques for Nonlinear Impairment Compensation using Neural Networks

March 1, 2022/Journal of Lightwave Technology

Neural networks (NNs) are attractive for nonlinear impairment compensation applications in communication systems, such as optical fiber nonlinearity, nonlinearity of driving amplifiers, and nonlinearity of semiconductor optical amplifiers. Without prior knowledge of the transmission link or the hardware

Zero-Shot Cross-Lingual Machine Reading Comprehension via Inter-Sentence Dependency Graph

March 1, 2022/Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI-2022)

We target the task of cross-lingual Machine Reading Comprehension (MRC) in the direct zero-shot setting, by incorporating syntactic features from Universal Dependencies (UD), and the key features we use are the syntactic relations within each sentence. While previous work has demonstrated effective syntax-guided

AI-Driven Applications over Telecom Networks by Distributed Fiber Optic Sensing Technologies

February 21, 2022/Photonics West 2022

By employing distributed fiber optic sensing (DFOS) technologies, field deployed fiber cables can be utilized as not only communication media for data transmissions but also sensing media for continuously monitoring of the physical phenomenon along the entire route. The fiber can be used to monitor ambient

Confidence and Dispersity Speak – Characterizing Prediction Matrix for Unsupervised Accuracy Estimation

February 2, 2022/arXiv

This work aims to assess how well a model performs under distribution shifts without using labels. While recent methods study prediction confidence, this work reports prediction dispersity is another informative cue. Confidence reflects whether the individual prediction is certain, dispersity indicates

A Dispersion Managed Phase Only Modulation 18 GHz Optoelectronic Oscillator

February 1, 2022/IEEE Photonics Technology Letters

In this manuscript, we propose and experimentally demonstrate a dispersion-controlled optoelectronic oscillator with phase only modulator at 18 GHz. The generated microwave signal has a phase noise of −108 dBc/Hz at 10 kHz offset frequency and the integrated timing jitter is calculated to be 16.2 fs

Ordinal Quadruplet: Retrieval of Missing Labels in Ordinal Time Series

January 24, 2022/arXiv

In this paper, we propose an ordered time series classification framework that is robust against missing classes in the training data, i.e., during testing we can prescribe classes that are missing during training. This framework relies on two main components: (1) our newly proposed ordinal quadruplet

Codebook Design for Composite Beamforming in Next generation mmWave Systems

January 24, 2022/arXiv

In pursuance of the unused spectrum in higher frequencies, millimeter wave (mmWave) bands have a pivotal role. However, the high path loss and poor scattering associated with mmWave communications highlight the necessity of employing effective beamforming techniques. In order to efficiently search for

Multi user Beam Alignment in Presence of Multi path

January 24, 2022/arXiv

To overcome the high path loss and the intense shadowing in millimeter wave (mmWave) communications, effective beamforming schemes are required which incorporate narrow beams with high beamforming gains. The mmWave channel consists of a few spatial clusters each associated with an angle of departure

AE-StyleGAN: Improved Training of Style-Based Auto-Encoders

January 4, 2022/WACV 2022

StyleGANs have shown impressive results on data generation and manipulation in recent years, thanks to its disentangled style latent space. A lot of efforts have been made in inverting a pretrained generator, where an encoder is trained ad hoc after the generator is trained in a two-stage fashion. In

SplitBrain: Hybrid Data and Model Parallel Deep Learning

January 3, 2022/arXiv

The recent success of deep learning applications has coincided with those widely available powerful computational resources for training sophisticated machine learning models with huge datasets. Nonetheless, training large models such as convolutional neural networks using model parallelism (as opposed

A Deep Generative Model for Molecule Optimization via One Fragment Modification

January 1, 2022/Nature Machine Intelligence

Molecule optimization is a critical step in drug development to improve the desired properties of drug candidates through chemical modification. We have developed a novel deep generative model, Modof, over molecular graphs for molecule optimization. Modof modifies a given molecule through the prediction

Distributed Fiber Sensor Network Using Telecom Cables as Sensing Media: Technology Advancements and Applications

January 1, 2022/Journal of Optical Communications and Networking

Distributed fiber optic sensing (DFOS) is a rapidly evolving field that allows the existing optical fiber infrastructure for telecommunications to be reused for wide-area sensing. Using the backscattering mechanisms of glass—which includes Rayleigh, Brillouin, and Raman backscatter—it is possible

Detection and Localization of Stationary Weights Hanging on Aerial Telecommunication Fibers using Distributed Acoustic Sensing

December 20, 2021/Optics Express

For the first time to our knowledge, a stationary weight hanging on an operational aerial telecommunication field fiber was detected and localized using only ambient data collected by a φ-DAS system. Although stationary weights do not create temporally varying signals, and hence cannot be observed directly

AQuA: Analytical Quality Assessment for Optimizing Video Analytics Systems

December 15, 2021/The Sixth ACM/IEEE Symposium on Edge Computing (SEC 2021)

Millions of cameras at edge are being deployed to power a variety of different deep learning applications. However, the frames captured by these cameras are not always pristine – they can be distorted due to lighting issues, sensor noise, compression etc. Such distortions not only deteriorate visual

Dynamic Causal Discovery in Imitation Learning

December 14, 2021/Causal Inference Challenges in Sequential Decision Making: Bridging Theory and Practice - A NeurIPS 2021 Workshop

Using deep reinforcement learning (DRL) to recover expert policies via imitation has been found to be promising in a wide range of applications. However, it remains a difficult task to interpret the control policy learned by the agent. Difficulties mainly come from two aspects: 1) agents in DRL are usually

Edge-based fever screening system over private 5G

December 14, 2021/The Sixth ACM/IEEE Symposium on Edge Computing (SEC 2021)

Edge computing and 5G have made it possible to perform analytics closer to the source of data and achieve super-low latency response times, which isn’t possible with centralized cloud deployment. In this paper, we present a novel fever screening system, which uses edge machine learning techniques and

InfoGCL: Information-Aware Graph Contrastive Learning

December 14, 2021/Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS 2021), Virtual-only Conference

InfoGCL: Information-Aware Graph Contrastive Learning Various graph contrastive learning models have been proposed to improve the performance of tasks on graph datasets in recent years. While effective and prevalent, these models are usually carefully customized. In particular, despite all recent work

Shaping mmWave Wireless Channel via Multi-Beam Design using Reconfigurable Intelligent Surfaces

December 7, 2021/IEEE Globecom - Workshop on Reconfigurable Intelligent Surfaces for Future Wireless Communications

Millimeter-wave (mmWave) communications is considered as a key enabler towards the realization of next-generation wireless networks, due to the abundance of available spectrum at mmWave frequencies. However, mmWave suffers from high free-space path-loss and poor scattering resulting in mostly line-of-sight

Magic-Pipe: Self-optimizing video analytics pipelines

December 6, 2021/Middleware 2021

Microservices-based video analytics pipelines routinely use multiple deep convolutional neural networks. We observe that the best allocation of resources to deep learning engines (or microservices) in a pipeline, and the best configuration of parameters for each engine vary over time, often at a timescale

SmartSlice: Dynamic, Self-optimization of Application’s QoS requests to 5G networks

December 6, 2021/The 5th International Symposium on 5G Emerging Technologies (5GET 2021)

Applications can tailor a network slice by specifying a variety of QoS attributes related to application-specific performance, function or operation. However, some QoS attributes like guaranteed bandwidth required by the application do vary over time. For example, network bandwidth needs of video streams

You Are What and Where You Are: Graph Enhanced Attention Network for Explainable POI Recommendation

November 15, 2021/30th ACM International Conference on Information and Knowledge Management (CIKM 2021)

Point-of-interest (POI) recommendation is an emerging area of research on location-based social networks to analyze user behaviors and contextual check-in information. For this problem, existing approaches, with shallow or deep architectures, have two major drawbacks. First, for these approaches, the

Boosting Cross-Lingual Transfer via Self-Learning with Uncertainty Estimation

November 11, 2021/The 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP 2021)

Recent multilingual pre-trained language models have achieved remarkable zero-shot performance, where the model is only finetuned on one source language and directly evaluated on target languages. In this work, we propose a self-learning framework that further utilizes unlabeled data of target languages,

Recommend for a Reason: Unlocking the Power of Unsupervised Aspect-Sentiment Co-Extraction

November 11, 2021/The 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP 2021)

Compliments and concerns in reviews are valuable for understanding users’ shopping interests and their opinions with respect to specific aspects of certain items. Existing review-based recommenders favor large and complex language encoders that can only learn latent and uninterpretable text representations.

Retrieval, Analogy, and Composition: A framework for Compositional Generalization in Image Captioning

November 7, 2021/EMNLP 2021

Image captioning systems are expected to have the ability to combine individual concepts when describing scenes with concept combinations that are not observed during training. In spite of significant progress in image captioning with the help of the autoregressive generation framework, current approaches

Team Papelo at FEVEROUS: Multi-hop Evidence Pursuit

November 7, 2021/The Fourth Workshop on Fast Extraction and Verificiation (FEVER)

We develop a system for the FEVEROUS fact extraction and verification task that ranks an initial set of potential evidence and then pursues missing evidence in subsequent hops by trying to generate it, with a “next hop prediction module” whose output is matched against page elements in a predicted

Interpreting Convolutional Sequence Model by Learning Local Prototypes with Adaptation Regularization

November 5, 2021/30th ACM International Conference on Information and Knowledge Management (CIKM 2021)

In many high-stakes applications of machine learning models, outputting only predictions or providing statistical confidence is usually insufficient to gain trust from end users, who often prefer a transparent reasoning paradigm. Despite the recent encouraging developments on deep networks for sequential

Structural Temporal Graph Neural Networks for Anomaly Detection in Dynamic Graphs

November 5, 2021/30th ACM International Conference on Information and Knowledge Management (CIKM 2021)

Detecting anomalies in dynamic graphs is a vital task, with numerous practical applications in areas such as security, finance, and social media. Existing network embedding based methods have mostly focused on learning good node representations, whereas largely ignoring the subgraph structural changes

Bipolar Cyclic Linear Coding for Brillouin Optical Time Domain Analysis

November 3, 2021/OFS 2020

We demonstrate, for the first time, that cyclic linear pulse coding can be bipolar for BOTDA sensors, breaking the unipolar limitation of linear coding techniques and elevating the coding gain for a given code length.

First Field Trial of Monitoring Vehicle Traffic on Multiple Routes by Using Photonic Switch and Distributed Fiber Optics Sensing System on Standard Telecom Networks

November 3, 2021/OFS 2021

We demonstrated for the first time that motor vehicle traffic and road capacity on multiple fiber routes can be monitored by using a distributed-fiber-optics-sensing system with a photonic switch on in-service telecom fiber cables.

Prediction of Non-Muscle Invasive Bladder Cancer Recurrence using Machine Learning of Quantitative Nuclear Features

October 29, 2021/Modern Pathology

Non-muscle invasive bladder cancer (NMIBC) generally has a good prognosis, however, recurrence after transurethral resection (TUR), the standard primary treatment, is a major problem. Clinical management after TUR has been based on risk classification using clinicopathological factors, but these classifications

CamTuner: Reinforcement Learning based System for Camera Parameter Tuning to enhance Analytics

October 26, 2021/arXiv

Video analytics systems critically rely on video cameras, which capture high quality video frames, to achieve high analytics accuracy. Although modern video cameras often expose tens of configurable parameter settings that can be set by end users, deployment of surveillance cameras today often uses a

Convolutional Transformer based Dual Discriminator Generative Adversarial Networks for Video Anomaly Detection

October 24, 2021/29th ACM International Conference on Multimedia (ACM Multimedia 2021)

Detecting abnormal activities in real-world surveillance videos is an important yet challenging task as the prior knowledge about video anomalies is usually limited or unavailable. Despite that many approaches have been developed to resolve this problem, few of them can capture the normal spatio-temporal

Dual Projection Generative Adversarial Networks for Conditional Image Generation

October 11, 2021/ICCV 2021

Conditional Generative Adversarial Networks (cGANs) extend the standard unconditional GAN framework to learning joint data-label distributions from samples, and have been established as powerful generative models capable of generating high-fidelity imagery. A challenge of training such a model lies in

Learning Cross-Modal Contrastive Features for Video Domain Adaptation

October 11, 2021/ICCV 2021, Virtual

Learning transferable and domain adaptive feature representations from videos is important for video-relevant tasks such as action recognition. Existing video domain adaptation methods mainly rely on adversarial feature alignment, which has been derived from the RGB image space. However, video data is

Learning Higher-order Object Interactions for Keypoint-based Video Understanding

October 11, 2021/SRVU - ICCV 2021 Workshop

Action recognition is an important problem that requires identifying actions in video by learning complex interactions across scene actors and objects. However, modern deep-learning based networks often require significant computation and may capture scene context using various modalities that further

Towards Robustness of Deep Neural Networks via Networks via Regularization

October 11, 2021/ICCV 2021

Recent studies have demonstrated the vulnerability of deep neural networks against adversarial examples. In-spired by the observation that adversarial examples often lie outside the natural image data manifold and the intrinsic dimension of image data is much smaller than its pixel space dimension, we

UAC: An Uncertainty-Aware Face Clustering Algorithm

October 11, 2021/IEEE/CVF International Conference on Computer Vision (ICCV) RLQ Workshop

We investigate ways to leverage uncertainty in face images to improve the quality of the face clusters. We observe that popular clustering algorithms do not produce better quality clusters when clustering probabilistic face representations that implicitly model uncertainty – these algorithms predict

Employing Telecom Fiber Cables as Sensing Media for Road Traffic Applications

October 7, 2021/WOCC 2021

Distributed fiber optic sensing systems (DFOS) allow deployed fiber cables to be sensing media, not only dedicated function of data transmission. The fiber cable can monitor the ambient environment over wide area for many applications. We review recent field trial results, and show how artificial intelligence

AppSlice: A system for application-centric design of 5G and edge computing applications

October 6, 2021/12th International Conference on Network of the Future (NoF 2021)

Applications that use edge computing and 5G to improve response times consume both compute and network resources. However, 5G networks manage only network resources without considering the application’s compute requirements, and container orchestration frameworks manage only compute resources without

A Silicon Photonic-Electronic Neural Network for Fiber Nonlinearity Compensation

October 1, 2021/Nature Electronics

In optical communication systems, fibre nonlinearity is the major obstacle in increasing the transmission capacity. Typically, digital signal processing techniques and hardware are used to deal with optical communication signals, but increasing speed and computational complexity create challenges for

DataX: A system for Data eXchange and transformation of streams

September 26, 2021/The 14th International Symposium on Intelligent Distributed Computing (IDC 2021)

The exponential growth in smart sensors and rapid progress in 5G networks is creating a world awash with data streams. However, a key barrier to building performant multi-sensor, distributed stream processing applications is high programming complexity. We propose DataX, a novel platform that improves

Guided Acoustic Brillouin Scattering Measurements In Optical Communication Fibers

September 1, 2021/Optics Express

Guided acoustic Brillouin (GAWBS) noise is measured using a novel, homodyne measurement technique for four commonly used fibers in long-distance optical transmission systems. The measurements are made with single spans and then shown to be consistent with separate multi-span long-distance measurements.

Optical Fiber Sensing Technology Visualizing the Real World via Network Infrastructures – AI technologies for traffic monitoring

August 30, 2021/The Institute of Electronics, Information and Communication Engineers (IEICE)

Optical fibers have a sensing function that captures environmental changes around the fiber cable. According to the recent technology evolution of optical transmission and AI, the application of the fiber sensing has expanded and visualization accuracy has improved. We have proposed to monitor the traffic

F3S: Free Flow Fever Screening

August 23, 2021/7th IEEE International Conference on Smart Computing (SMARTCOMP 2021)

Identification of people with elevated body temperature can reduce or dramatically slow down the spread of infectious diseases like COVID-19. We present a novel fever-screening system, F 3 S, that uses edge machine learning techniques to accurately measure core body temperatures of multiple individuals

Domain oriented Language Modeling with Adaptive Hybrid Masking and Optimal Transport Alignment

August 18, 2021/KDD 2021: ACM SIGKDD Conference on Knowledge Discovery and Data Mining, SIGKDD 2021

Motivated by the success of pre-trained language models such as BERT in a broad range of natural language processing (NLP) tasks, recent research efforts have been made for adapting these models for different application domains. Along this line, existing domain-oriented models have primarily followed

Multi-Scale One-Class Recurrent Neural Networks for Discrete Event Sequence Anomaly Detection

August 18, 2021/ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2021)

Discrete event sequences are ubiquitous, such as an ordered event series of process interactions in Information and Communication Technology systems. Recent years have witnessed increasing efforts in detecting anomalies with discrete event sequences. However, it remains an extremely difficult task due

SIGL: Securing Software Installations Through Deep Graph Learning

August 13, 2021/USENIX Security 2021 - The 30th USENIX Security Symposium

Many users implicitly assume that software can only be exploited after it is installed. However, recent supply-chain attacks demonstrate that application integrity must be ensured during installation itself. We introduce SIGL, a new tool for detecting malicious behavior during software installation.

Overcoming Poor Word Embeddings with Word Definitions

August 5, 2021/SEM 2021 Workshop at ACL-IJCNLP 2021

Modern natural language understanding models depend on pretrained subword embeddings, but applications may need to reason about words that were never or rarely seen during pretraining. We show that examples that depend critically on a rarer word are more challenging for natural language inference models.

SkyHAUL: A Self-Organizing Gigabit Network In The Sky

July 26, 2021/ACM Mobihoc 2021

We design and build SkyHaul, the first large-scale, self-organizing network of Unmanned Aerial Vehicles (UAVs) that are connected using a mm Wave wireless mesh backhaul. While the use of a mmWave backhaul paves the way for a new class of bandwidth-intensive, latency-sensitive cooperative applications

DECODE: A Deep-learning Framework for Condensing Enhancers and Refining Boundaries with Large-scale Functional Assays

July 25, 2021/ISMB/ECCB 2021

MotivationMapping distal regulatory elements, such as enhancers, is a cornerstone for elucidating how genetic variations may influence diseases. Previous enhancer-prediction methods have used either unsupervised approaches or supervised methods with limited training data. Moreover, past approaches have

Hierarchical Imitation Learning with Contextual Bandits for Dynamic Treatment Regimes

July 24, 2021/The Thirty-eighth International Conference on Machine Learning (ICML 2021)

Imitation learning has been proved to be effective in mimicking experts’ behaviors from their demonstrations without access to explicit reward signals. Meanwhile, complex tasks, e.g., dynamic treatment regimes for patients with comorbidities, often suggest significant variability in expert demonstrations

On Single-User Interactive Beam Alignment in Millimeter Wave Systems: Impact of Feedback Delay

July 12, 2021/The IEEE International Symposium on Information Theory (IEEE ISIT 2021)

Narrow beams are key to wireless communications in millimeter wave frequency bands. Beam alignment (BA) allows the base station (BS) to adjust the direction and width of the beam used for communication. During BA, the BS transmits a number of scanning beams covering different angular regions. The goal

An Efficient Approach for Placing Distributed Fiber Optic Sensors with Concurrent Sensing Capability

July 6, 2021/OECC 2021

We propose an efficient approach for placing distributed fiber optic sensors (DFOS) with concurrent sensing capability. It consumes 5.7% to 9.5% fewer sensors than that using DFOS without concurrent sensing, for covering the same network.

Field Trial of Cable Safety Protection and Road Traffic Monitoring over Operational 5G Transport Network with Fiber Sensing and On-Premise AI Technologies

July 6, 2021/OECC 2021 - PDP

We report the distributed-fiber-sensing field trial results over a 5G-transport-network. A standard communication fiber is used with real-time AI processing for cable self-protection, cable-cut threat assessment and road traffic monitoring in a long-term continuous test.

FACESEC: A Fine-grained Robustness Evaluation Framework for Face Recognition Systems

June 25, 2021/CVPR 2021 - IEEE/CVF Conference on Computer Vision and Pattern Recognition

We present FACESEC, a framework for fine-grained robustness evaluation of face recognition systems. FACESEC evaluation is performed along four dimensions of adversarial modeling: the nature of perturbation (e.g., pixel-level or face accessories), the attacker’s system knowledge (about training data

SpaceBeam: LiDAR-Driven One-Shot mmWave Beam Management

June 24, 2021/19th ACM International Conference on Mobile Systems, Applications, and Services (MobiSys 2021)

mmWave 5G networks promise to enable a new generation of networked applications requiring a combination of high throughput and ultra-low latency. However, in practice, mmWave performance scales poorly for large numbers of users due to the significant overhead required to manage the highly-directional

Cross-Domain Similarity Learning for Face Recognition in Unseen Domains

June 19, 2021/CVPR 2021, Virtual

Face recognition models trained under the assumption of identical training and test distributions often suffer from poor generalization when faced with unknown variations, such as a novel ethnicity or unpredictable individual make-ups during test time. In this paper, we introduce a novel cross-domain

Divide-and-Conquer for Lane-Aware Diverse Trajectory Prediction

June 19, 2021/CVPR 2021, Virtual

Trajectory prediction is a safety-critical tool for autonomous vehicles to plan and execute actions. Our work addresses two key challenges in trajectory prediction, learning multimodal outputs, and better predictions by imposing constraints using driving knowledge. Recent methods have achieved strong

Fusing the Old with the New: Learning Relative Pose with Geometry-Guided Uncertainty

June 19, 2021/CVPR 2021, Virtual

Learning methods for relative camera pose estimation have been developed largely in isolation from classical geometric approaches. The question of how to integrate predictions from deep neural networks (DNNs) and solutions from geometric solvers, such as the 5-point algorithm [37], has as yet remained

Automated Anomaly Detection via Curiosity-Guided Search and Self-Imitation Learning

June 15, 2021/The IEEE Transactions on Neural Networks and Learning Systems

Anomaly detection is an important data mining task with numerous applications, such as intrusion detection, credit card fraud detection, and video surveillance. However, given a specific complicated task with complicated data, the process of building an effective deep learning-based system for anomaly

Survivable Distributed Fiber Optic Sensors Placement against Single Link Failure

June 14, 2021/IEEE International Conference on Communications (IEEE ICC 2021)

Empowered by the rapid advancement of fiber optic sensing techniques in recent years, network carriers are able to upgrade their network infrastructure beyond the basic communication services with extra sensing applications and services (e.g., monitoring traffic and road condition, leakage detection,

Unsupervised Concept Representation Learning for Length-Varying Text Similarity

June 11, 2021/NAACL 2021 – 2021 Annual Conference of the North American Chapter

Measuring document similarity plays an important role in natural language processing tasks. Most existing document similarity approaches suffer from the information gap caused by context and vocabulary mismatches when comparing varying-length texts. In this paper, we propose an unsupervised concept representation

Estimation of Core-Cladding Concentricity Error From GAWBS Noise Spectrum

June 7, 2021/OFC 2021

CCCE in a 60-km fiber is estimated from its GAWBS noise spectrum by comparing the TR 1m modes with the R 0m modes. The estimated CCCE value 0.73 μm is consistent with conventional measurements of 0.6–0.8 μm.

Field Trial of Abnormal Activity Detection and Threat Level Assessment with Fiber Optic Sensing for Telecom Infrastructure Protection

June 7, 2021/OFC 2021

We report the field trial results of monitoring abnormal activities near deployed cable with fiber-optic-sensing technology for cable protection. Detection and position determination of abnormal events and evaluating the threat to the cable is realized.

Nonlinear Impairment Compensation using Neural Networks

June 7, 2021/OFC 2021

Neural networks are attractive for nonlinear impairment compensation applications in communication systems. In this paper, several approaches to reduce computational complexity of the neural network-based algorithms are presented.

Static Weight Detection and Localization on Aerial Fiber Cables using Distributed Acoustic Sensing

June 7, 2021/OFC 2021

We demonstrated for the first time to our knowledge, the detection and localization of a static weight on an aerial cable by using frequency domain decomposition analysis of ambient vibrations detected by a φ-DAS system.

Vehicle Run-Off-Road Event Automatic Detection by Fiber Sensing Technology

June 7, 2021/OFC 2021

We demonstrate a new application of fiber-optic-sensing and machine learning techniques for vehicle run-off-road events detection to enhance roadway safety and efficiency. The proposed approach achieves high accuracy in a testbed under various experimental conditions.

Automatic Fine-Grained Localization of Utility Pole Landmarks on Distributed Acoustic Sensing Traces Based on Bilinear Resnets

June 6, 2021/2021 IEEE International Conference on Acoustics, Speech and Signal Processing

In distributed acoustic sensing (DAS) on aerial fiber-optic cables, utility pole localization is a prerequisite for any subsequent event detection. Currently, localizing the utility poles on DAS traces relies on human experts who manually label the poles’ locations by examining DAS signal patterns

Distributed Fiber Sensor Network using Telecom Cables as Sensing Media: Applications

June 6, 2021/OFC 2021

Distributed fiber optical systems (DFOS) allow deployed optical cables to monitor the ambient environment over wide geographic area. We review recent field trial results, and show how DFOS can be made compatible with passive optical networks (PONs).

Field Trial of Vibration Detection and Localization using Coherent Telecom Transponders over 380-km Link

June 6, 2021/OFC 2021, Virtual

We demonstrate vibration detection and localization based on extracting optical phase from the DSP elements of a coherent receiver in bidirectional WDM transmission of 200-Gb/s DP-16QAM over 380 km of installed field fiber.

Optics and Biometrics

May 31, 2021/Optics & Photonics News

Forget passwords—identity verification can now be accomplished with the touch of a finger or in the blink of an eye as the biometrics field expands to encompass new techniques and application areas.

ECO: Edge-Cloud Optimization of 5G applications

May 10, 2021/The 21st IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGrid 2021), Melbourne, Victoria, Australia

Centralized cloud computing with 100+ milliseconds network latencies cannot meet the tens of milliseconds to sub-millisecond response times required for emerging 5G applications like autonomous driving, smart manufacturing, tactile internet, and augmented or virtual reality. We describe a new, dynamic

Disentangled Recurrent Wasserstein Auto-Encoder

May 4, 2021/ICLR 2021

Learning disentangled representations leads to interpretable models and facilitates data generation with style transfer, which has been extensively studied on static data such as images in an unsupervised learning framework. However, only a few works have explored unsupervised disentangled sequential

Hopper: Multi-hop Transformer for Spatio-Temporal Reasoning

May 4, 2021/ICLR 2021

This paper considers the problem of spatiotemporal object-centric reasoning in videos. Central to our approach is the notion of object permanence, i.e., the ability to reason about the location of objects as they move through the video while being occluded, contained or carried by other objects. Existing

Deep Multi-Instance Contrastive Learning with Dual Attention for Anomaly Precursor Detection

May 1, 2021/SIAM International Conference on Data Mining, Virtual Conference (SDM21)

Prognostics or early detection of incipient faults by leveraging the monitoring time series data in complex systems is valuable to automatic system management and predictive maintenance. However, this task is challenging. First, learning the multi-dimensional heterogeneous time series data with various

Ranking-based Convolutional Neural Network Models for Peptide-MHC Binding Prediction

May 1, 2021/Frontiers in Molecular Biosciences - Biological Modeling and Simulation

T-cell receptors can recognize foreign peptides bound to major histocompatibility complex (MHC) class-I proteins, and thus trigger the adaptive immune response. Therefore, identifying peptides that can bind to MHC class-I molecules plays a vital role in the design of peptide vaccines. Many computational

AutoOD: Neural Architecture Search for Outlier Detection

April 23, 2021/ICDE 2021 - The 37th IEEE International Conference on Data Engineering

Outlier detection is an important data mining task with numerous applications such as intrusion detection, credit card fraud detection, and video surveillance. However, given a specific task with complex data, the process of building an effective deep learning based system for outlier detection still

Learning to Drop: Robust Graph Neural Network via Topological Denoising

March 12, 2021/WSDM 2021 - The 14th ACM International WSDM Conference on Web Seach and Data Mining

Graph Neural Networks (GNNs) have shown to be powerful tools for graph analytics. The key idea is to recursively propagate and aggregate information along the edges of the given graph. Despite their success, however, the existing GNNs are usually sensitive to the quality of the input graph. Real-world

Multi-Task Recurrent Modular Networks

March 9, 2021/AAAI 2021 - 35th AAAI Conference on Artificial Intelligence

We consider the models of deep multi-task learning with recurrent architectures that exploit regularities across tasks to improve the performance of multiple sequence processing tasks jointly. Most existing architectures are painstakingly customized to learn task relationships for different problems,

Dynamic Gaussian Mixture based Deep Generative Model For Robust Forecasting on Sparse Multivariate Time Series

February 22, 2021/AAAI 2021 - 35th AAAI Conference on Artificial Intelligence

Forecasting on Sparse Multivariate Time Series Forecasting on sparse multivariate time series (MTS) aims to model the predictors of future values of time series given their incomplete past, which is important for many emerging applications. However, most existing methods process MTS’s individually,

A Multi-Scale Conditional Deep Model for Tumor Cell Ratio Counting

February 17, 2021/SPIE Medical Imaging 2021: MI109: Digital and Computational Pathology

We propose a method to accurately obtain the ratio of tumor cells over an entire histological slide. We use deep fully convolutional neural network models trained to detect and classify cells on images of H&E-stained tissue sections. Pathologists’ labels consisting of exhaustive nuclei locations and

Improving neural network robustness through neighborhood preserving layers

January 15, 2021/Manifold Learning from Euclid to Riemann: Workshop at ICPR 2021

One major source of vulnerability of neural nets in classification tasks is from overparameterized fully connected layers near the end of the network. In this paper, we propose a new neighborhood preserving layer which can replace these fully connected layers to improve the network robustness. Networks

Cross-Modality 3D Object Detection

January 5, 2021/WACV 2021, Virtual

In this paper, we focus on exploring the fusion of images and point clouds for 3D object detection in view of the complementary nature of the two modalities, i.e., images possess more semantic information while point clouds specialize in distance sensing. To this end, we present a novel two-stage multi-modal

Set Augmented Triplet Loss for Video Person Re-Identification

January 5, 2021/WACV 2021, Virtual

Modern video person re-identification (re-ID) machines are often trained using a metric learning approach, supervised by a triplet loss. The triplet loss used in video re-ID is usually based on so-called clip features, each aggregated from a few frame features. In this paper, we propose to model the

Parameterized Explainer for Graph Neural Network

December 12, 2020/Thirty-Fourth Annual Conference on Neural Information Processing Systems (NeurIPS 2020)

Despite recent progress in Graph Neural Networks (GNNs), explaining predictions made by GNNs remains a challenging open problem. The leading method independently addresses the local explanations (i.e., important subgraph structure and node features) to interpret why a GNN model makes the prediction for

This is Why We Can’t Cache Nice Things: Lightning-Fast Threat Hunting using Suspicion-Based Hierarchical Storage

December 11, 2020/2020 Annual Computer Security Applications Conference

Recent advances in causal analysis can accelerate incident response time, but only after a causal graph of the attack has been constructed. Unfortunately, existing causal graph generation techniques are mainly offline and may take hours or days to respond to investigator queries, creating greater opportunity

Field Trial of Distributed Fiber Sensor Network Using Operational Telecom Fiber Cables as Sensing Media

December 10, 2020/ECOC 2020

We demonstrate fiber optic sensing systems in a distributed fiber sensor network built on existing telecom infrastructure to detect temperature, acoustic effects, vehicle traffic, etc. Measurements are also demonstrated with different network topologies and simultaneously sensing four fiber routes with

Address Challenges in Placing Distributed Fiber Optic Sensors

December 9, 2020/ECOC 2020

We are the first to investigate a novel problem, called distributed fiber optic sensor placement, in the context of Infrastructure-as-a-Sensor. We propose an ILP-based optimal solution and a close-to-optimal heuristic solution, both of which aim at minimizing the cost of sensors.

Channel Recurrent Attention Networks for Video Pedestrian Retrieval

November 30, 2020/ACCV 2020, Kyoto, Japan

Full attention, which generates an attention value per element of the input feature maps, has been successfully demonstrated to be beneficial in visual tasks. In this work, we propose a fully attentional network, termed channel recurrent attention network, for the task of video pedestrian retrieval.

T2-Net: A Semi-supervised Deep Model for Turbulence Forecasting

November 20, 2020/ICDM 2020 - The 20th IEEE International Conference on Data Mining

Accurate air turbulence forecasting can help airlines avoid hazardous turbulence, guide the routes that keep passengers safe, maximize efficiency, and reduce costs. Traditional turbulence forecasting approaches heavily rely on painstakingly customized turbulence indexes, which are less effective in dynamic

Uncertainty Aware Physically Guided Proxy Tasks for Unseen Domain Face Anti-Spoofing

November 20, 2020/arXiv

Face anti-spoofing (FAS) seeks to discriminate genuine faces from fake ones arising from any type of spoofing attack. Due to the wide variety of attacks, it is implausible to obtain training data that spans all attack types. We propose to leverage physical cues to attain better generalization on unseen

Multi-user Beam Alignment for Millimeter Wave Systems in Multi-path Environments

November 1, 2020/54th Annual Asilomar Conference on Signals, Systems, and Computers

Directional transmission patterns (a.k.a. narrow beams) are the key to wireless communications in millimeter wave (mmWave) frequency bands which suffer from high path loss, severe shadowing, and intense blockage. In addition, the propagation channel in mmWave frequencies incorporates only a few number

Anomaly Detection on Web-User Behaviors through Deep Learning

October 23, 2020/16th EAI International Conference on Security and Privacy in Communication Networks (SecureComm 2020)

The modern Internet has witnessed the proliferation of web applications that play a crucial role in the branding process among enterprises. Web applications provide a communication channel between potential customers and business products. However, web applications are also targeted by attackers due

VESSELS: Efficient and Scalable Deep Learning Prediction on Trusted Processors

October 21, 2020/ACM Symposium on Cloud Computing 2020 (SoCC 2020)

Deep learning systems on the cloud are increasingly targeted by attacks that attempt to steal sensitive data. Intel SGX has been proven effective to protect the confidentiality and integrity of such data during computation. However, state-of-the-art SGX systems still suffer from substantial performance

Voting Based Approaches For Differentially Private Federated Learning

October 6, 2020/arXiv

Differentially Private Federated Learning (DPFL) is an emerging field with many applications. Gradient averaging-based DPFL methods require costly communication rounds and hardly work with large capacity models due to the explicit dimension dependence in its added noise. In this work, inspired by knowledge

New Methods for Non-Destructive Underground Fiber Localization using Distributed Fiber Optic Sensing Technology

October 5, 2020/OECC 2020

To the best of our knowledge, we present the first underground fiber cable position detection methods using distributed fiber optic sensing (DFOS) technology. Meter level localization accuracy is achieved in the results.

3D Finger Vein Biometric Authentication with Photoacoustic Tomography

October 1, 2020/Applied Optics

Biometric authentication is the recognition of human identity via unique anatomical features. The development of novel methods parallels widespread application by consumer devices, law enforcement, and access control. In particular, methods based on finger veins, as compared to face and fingerprints,

Anomalous Event Sequence Detection

September 24, 2020/IEEE Intelligent Systems

Anomaly detection has been widely applied in modern data-driven security applications to detect abnormal events/entities that deviate from the majority. However, less work has been done in terms of detecting suspicious event sequences/paths, which are better discriminators than single events/entities

Redefining Passive in Backscattering with Commodity Devices

September 21, 2020/The 26th Annual International Conference on Mobile Computing and Networking (MobiCom 2020)

The recent innovation of frequency-shifted (FS) backscatter allows for backscattering with commodity devices, which are inherently half-duplex. However, their reliance on oscillators for generating the frequency-shifting signal on the tag, forces them to incur the transient phase of the oscillator before

RFGo: A Seamless Self-checkout System for Apparel Stores Using RFID

September 21, 2020/The 26th Annual International Conference on Mobile Computing and Networking (MobiCom 2020)

Retailers are aiming to enhance customer experience by automating the checkout process. The key impediment here is the effort to manually align the product barcode with the scanner, requiring sequential handling of items without blocking the line-of-sight of the laser beam. While recent systems such

Node Classification in Temporal Graphs through Stochastic Sparsification and Temporal Structural Convolution

September 18, 2020/ECML-PKDD 2020 - The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases

Node classification in temporal graphs aims to predict node labels based on historical observations. In real-world applications, temporal graphs are complex with both graph topology and node attributes evolving rapidly, which poses a high overfitting risk to existing graph learning approaches. In this

Prediction of Early Recurrence of Hepatocellular Carcinoma after Resection using Digital Pathology Images Assessed by Machine Learning

September 18, 2020/Modern Pathology

Hepatocellular carcinoma (HCC) is a representative primary liver cancer caused by long-term and repetitive liver injury. Surgical resection is generally selected as the radical cure treatment. Because the early recurrence of HCC after resection is associated with low overall survival, the prediction

Model-Based Autoencoders for Imputing Discrete single-cell RNA-seq Data

September 16, 2020/Methods (Elsevier)

Deep neural networks have been widely applied for missing data imputation. However, most existing studies have been focused on imputing continuous data, while discrete data imputation is under-explored. Discrete data is common in real world, especially in research areas of bioinformatics, genetics, and

Anti-spoofing Face Recognition Using Infrared Structure Light

September 13, 2020/2019 Frontiers in Optics/Laser Science Conference

We demonstrate an anti-spoofing face recognition system that is able to differentiate real human face with 3D printed materials. Face images captured in infrared structure light are analyzed for surface materials and spatial structure.

Tripping Through Time: Efficient Localization of Activities in Videos

September 11, 2020/BMVC 2020

Localizing moments in untrimmed videos via language queries is a new and interesting task that requires the ability to accurately ground language into video. Previous works have approached this task by processing the entire video, often more than once, to localize relevant activities. In the real world

Adaptation Across Extreme Variations using Unlabeled Bridges

September 7, 2020/BMVC’20, Manchester, UK

We tackle an unsupervised domain adaptation problem for which the domain discrepancy between labeled source and unlabeled target domains is large, due to many factors of inter- and intra-domain variation. While deep domain adaptation methods have been realized by reducing the domain discrepancy, these

Pseudo RGB-D for Self-Improving Monocular SLAM and Depth Prediction

August 28, 2020/ECCV 2020 - The 16th European Conference on Computer Vision, Glasgow, UK

Classical monocular Simultaneous Localization And Mapping (SLAM) and the recently emerging convolutional neural networks (CNNs) for monocular depth prediction represent two largely disjoint approaches towards building a 3D map of the surrounding environment. In this paper, we demonstrate that the coupling

Domain Adaptive Semantic Segmentation using Weak Labels

August 23, 2020/ECCV 2020 - The 16th European Conference on Computer Vision, Glasgow, UK

We propose a novel framework for domain adaptation in semantic segmentation with image-level weak labels in the target domain. The weak labels may be obtained based on a model prediction for unsupervised domain adaptation (UDA), or from a human oracle in a new weakly-supervised domain adaptation (WDA)

Image Stitching and Rectification for Hand-Held Cameras

August 23, 2020/ECCV 2020 - The 16th European Conference on Computer Vision, Glasgow, UK

In this paper, we derive a new differential homography that can account for the scanline-varying camera poses in Rolling Shutter (RS) cameras, and demonstrate its application to carry out RS-aware image stitching and rectification at one stroke. Despite the high complexity of RS geometry, we focus in

Improving Face Recognition by Clustering Unlabeled Faces in the Wild

August 23, 2020/ECCV 2020 - The 16th European Conference on Computer Vision, Glasgow, UK

While deep face recognition has benefited significantly from large-scale labeled data, current research is focused on leveraging unlabeled data to further boost performance, reducing the cost of human annotation. Prior work has mostly been in controlled settings, where the labeled and unlabeled data

Learning Monocular Visual Odometry via Self-Supervised Long-Term Modeling

August 23, 2020/ECCV 2020 - The 16th European Conference on Computer Vision, Glasgow, UK

Monocular visual odometry (VO) suffers severely from error accumulation during frame-to-frame pose estimation. In this paper, we present a self-supervised learning method for VO with special consideration for consistency over longer sequences. To this end, we model the long-term dependency in pose prediction

Learning to Optimize Domain Specific Normalization for Domain Generalization

August 23, 2020/ECCV 2020 – The 16th European Conference on Computer Vision, Glasgow, UK

We propose a simple but effective multi-source domain generalization technique based on deep neural networks by incorporating optimized normalization layers that are specific to individual domains. Our approach employs multiple normalization methods while learning separate affine parameters per domain.

Object Detection with a Unified Label Space from Multiple Datasets

August 23, 2020/ECCV 2020 - The 16th European Conference on Computer Vision, Glasgow, UK

Given multiple datasets with different label spaces, the goal of this work is to train a single object detector predicting over the union of all the label spaces. The practical benefits of such an object detector are obvious and significant—application-relevant categories can be picked and merged form

Shuffle and Attend: Video Domain Adaptation

August 23, 2020/ECCV 2020 - The 16th European Conference on Computer Vision, Glasgow, UK

We address the problem of domain adaptation in videos for the task of human action recognition. Inspired by image-based domain adaptation, we can perform video adaptation by aligning the features of frames or clips of source and target videos. However, equally aligning all clips is sub-optimal as not

SMART: Simultaneous Multi-Agent Recurrent Trajectory Prediction

August 23, 2020/ECCV 2020 - The 16th European Conference on Computer Vision, Glasgow, UK

We propose advances that address two key challenges in future trajectory prediction: (i) multimodality in both training data and predictions and (ii) constant time inference regardless of number of agents. Existing trajectory predictions are fundamentally limited by lack of diversity in training data,

BAFFLE: Decentralized Blockchain based Aggregator-Free Federated Learning

August 6, 2020/Blockchain 2020 - The 3rd IEEE International Conference on Blockchain, Rhodes Island, Greece

A key aspect of Federated Learning (FL) is the requirement of a centralized aggregator to maintain and update the global model. However, in many cases orchestrating a centralized aggregator might be infeasible due to numerous operational constraints. In this paper, we introduce BAFFLE, an aggregator

Stochastic Decision-Making Model for Aggregation of Residential Units with PV-Systems and Storages

August 2, 2020/The 2020 IEEE PES General Meeting, Montreal, Canada

Many residential energy consumers have installed photovoltaic (PV) panels and energy storage systems. These residential users can aggregate and participate in the energy markets. A stochastic decision making model for an aggregation of these residential units for participation in two-settlement markets

Robust Graph Representation Learning via Neural Sparsification

July 18, 2020/The 37th International Conference on Machine Learning (ICML 2020)

Graph representation learning serves as the core of important prediction tasks, ranging from product recommendation to fraud detection. Reallife graphs usually have complex information in the local neighborhood, where each node is described by a rich set of features and connects to dozens or even hundreds

Austere Flash Caching with Deduplication and Compression

July 15, 2020/2020 USENIX Annual Technical Conference (USENIX ATC '20)

Modern storage systems leverage flash caching to boost I/O performance, and enhancing the space efficiency and endurance of flash caching remains a critical yet challenging issue in the face of ever-growing data-intensive workloads. Deduplication and compression are promising data reduction techniques

Improving Face Recognition by Clustering Unlabeled Faces in the Wild (arXiv)

July 10, 2020

Read Improving Face Recognition by Clustering Unlabeled Faces in the Wild (arXiv). While deep face recognition has benefited significantly from large scale labeled data, current research is focused on leveraging unlabeled data to further boost performance, reducing the cost of human annotation. Prior

DeepTrack: Grouping RFID Tags Based on Spatio-temporal Proximity in Retail Spaces

July 6, 2020/IEEE International Conference on Computer Communications (IEEE Infocom 2020)

RFID applications for taking inventory and processing transactions in point-of-sale (POS) systems improve operational efficiency but are not designed to provide insights about customers’ interactions with products. We bridge this gap by solving the proximity grouping problem to identify groups of RFID

Improving Disentangled Text Representation Learning with Information Theoretical Guidance

July 5, 2020/ACL 2020

Learning disentangled representations of natural language is essential for many NLP tasks, e.g., conditional text generation, style transfer, personalized dialogue systems, etc. Similar problems have been studied extensively for other forms of data, such as images and videos. However, the discrete nature

On Optimal Multi-user Beam Alignment in Millimeter Wave Wireless Systems

June 21, 2020/2020 IEEE International Symposium on Information Theory (ISIT 2020)

Directional transmission patterns (a.k.a. narrow beams) are the key to wireless communications in millimeter wave (mmWave) frequency bands which suffer from high path loss and severe shadowing. In addition, the propagation channel in mmWave frequencies incorporates only a few number of spatial clusters

Peek-a-boo: Occlusion Reasoning in Indoor Scenes with Plane Representations

June 16, 2020/CVPR 2020

We address the challenging task of occlusion-aware indoor 3D scene understanding. We represent scenes by a set of planes, where each one is defined by its normal, offset and two masks outlining (i) the extent of the visible part and (ii) the full region that consists of both visible and occluded parts

Private-kNN Practical Differential Privacy for Computer Vision

June 16, 2020/CVPR 2020

With increasing ethical and legal concerns on privacy for deep models in visual recognition, differential privacy has emerged as a mechanism to disguise membership of sensitive data in training datasets. Recent methods like Private Aggregation of Teacher Ensembles (PATE) leverage a large ensemble of

Towards Universal Representation Learning for Deep Face Recognition

June 16, 2020/CVPR 2020

Recognizing wild faces is extremely hard as they appear with all kinds of variations. Traditional methods either train with specifically annotated variation data from target domains, or by introducing unlabeled target variation data to adapt from the training data. Instead, we propose a universal representation

Understanding Road Layout from Videos as a Whole

June 16, 2020/CVPR 2020

In this paper, we address the problem of inferring the layout of complex road scenes from video sequences. To this end, we formulate it as a top-view road attributes prediction problem and our goal is to predict these attributes for each frame both accurately and consistently. In contrast to prior work,

15 Keypoints Is All You Need

June 14, 2020/CVPR 2020

Pose-tracking is an important problem that requires identifying unique human pose-instances and matching them temporally across different frames in a video. However, existing pose-tracking methods are unable to accurately model temporal relationships and require significant computation, often computing

S3VAE: Self-Supervised Sequential VAE for Representation Disentanglement and Data Generation

June 14, 2020/CVPR 2020

We propose a sequential variational autoencoder to learn disentangled representations of sequential data (e.g., videos and audios) under self-supervision. Specifically, we exploit the benefits of some readily accessible supervision signals from input data itself or some off-the-shelf functional models

At the Speed of Sound: Efficient Audio Scene Classification

June 11, 2020/The Annual ACM International Conference on Multimedia Retrieval (ICMR 2020)

Efficient audio scene classification is essential for smart sensing platforms such as robots, medical monitoring, surveillance, or autonomous vehicles. We propose a retrieval-based scene classification architecture that combines recurrent neural networks and attention to compute embeddings for short

RULENet: End-to-end Learning with the Dual-estimator for Remaining Useful Life Estimation

June 10, 2020/2020 IEEE International Conference on Prognostics and Health Management, Detroit, MI

Remaining Useful Life (RUL) estimation is a key element in Predictive maintenance. System agnostic approaches which just utilize sensor and operational time series have gained popularity due to its ease of implementation. Due to the nature of measurement or degradation mechanisms, its accurate estimation

Chemical profiling of red wines using surface-1 enhanced Raman spectroscopy (SERS)

May 1, 2020/Analytical Methods

In this study, we explored surface-enhanced Raman spectroscopy (SERS) for analyzing red wine through several facile sample preparations. These approaches involved the direct analysis of red wine with Raman spectroscopy and the direct incubation of red wine with silver nanoparticles (i.e., AgNPs) and

Inductive and Unsupervised Representation Learning on Graph Structured Objects

April 30, 2020/8th International Conference on Learning Representations (ICLR 2020)

Inductive and unsupervised graph learning is a critical technique for predictive or information retrieval tasks where label information is difficult to obtain. It is also challenging to make graph learning inductive and unsupervised at the same time, as learning processes guided by reconstruction error

A Generic Edge-Empowered Graph Convolutional Network via Node-Edge Mutual Enhancement

April 24, 2020/The Web Conference 2020 (WWW 2020)

Graph Convolutional Networks (GCNs) have shown to be a powerful tool for analyzing graph-structured data. Most of previous GCN methods focus on learning a good node representation by aggregating the representations of neighboring nodes, whereas largely ignoring the edge information. Although few recent

Adversarial Cooperative Imitation Learning for Dynamic Treatment Regimes

April 24, 2020/The Web Conference 2020 (WWW 2020)

Recent developments in discovering dynamic treatment regimes (DTRs) have heightened the importance of deep reinforcement learning (DRL) which are used to recover the doctor’s treatment policies. However, existing DRL-based methods expose the following limitations: 1) supervised methods based on behavior

APTrace: A Responsive System for Agile Enterprise Level Causality Analysis

April 24, 2020/36th IEEE International Conference on Data Engineering (ICDE 2020)

While backtracking analysis has been successful in assisting the investigation of complex security attacks, it faces a critical dependency explosion problem. To address this problem, security analysts currently need to tune backtracking analysis manually with different case-specific heuristics. However,

Generating Followup Questions for Interpretable Multi hop Question Answering

March 31, 2020/arXiv

We propose a framework for answering open domain multi hop questions in which partial information is read and used to generate followup questions, to finally be answered by a pretrained single hop answer extractor. This framework makes each hop interpretable, and makes the retrieval associated with later

You Are What You Do: Hunting Stealthy Malware via Data Provenance Analysis

March 9, 2020/NDSS Symposium 2020

To subvert recent advances in perimeter and host security, the attacker community has developed and employed various attack vectors to make malware much more stealthy than before to penetrate the target system and prolong its presence. The advanced malware, or stealthy malware, impersonates or abuses

Demonstration of photonic neural network for fiber nonlinearity compensation in long-haul transmission systems

March 8, 2020/OFC 2020

We demonstrate the experimental implementation of photonic neural network for fiber nonlinearity compensation over a 10,080 km trans-pacific transmission link. Q-factor improvement of 0.51 dB is achieved with only 0.06 dB lower than numerical simulations.

First Proof That Geographic Location on Deployed Fiber Cable Can Be Determined by Using OTDR Distance Based on Distributed Fiber Optical Sensing Technology

March 8, 2020/OFC 2020

We demonstrated for the first time that geographic locations on deployed fiber cables can be determined accurately by using OTDR distances. The method involves vibration stimulation near deployed cables and distributed fiber optical sensing technology.

More Than Communications: Environment Monitoring Using Existing Data Center Network Infrastructure

March 8, 2020/OFC 2020

We propose reusing existing optical cables in metropolitan networks for distributed sensing using a bidirectional, dual-band architecture where communications and sensing signals can coexist with weak interaction on the same optical fiber.

Simultaneous Optical Fiber Sensing and Mobile Front-Haul Access over a Passive Optical Network

March 8, 2020/OFC 2020

We demonstrate a passive optical network (PON) that employs reflective semiconductor optical amplifiers (RSOAs) at optical network units (ONUs) to allow simultaneous data transmission with distributed fiber-optic sensing (DFOS) on individual distribution fibers.

Active Adversarial Domain Adaptation

March 2, 2020/WACV 2020, Snowmass Village, CO USA

We propose an active learning approach for transferring representations across domains. Our approach, active adversarial domain adaptation (AADA), explores a duality between two related problems: adversarial domain alignment and importance sampling for adapting models across domains. The former uses

Coordinated Joint Multimodal Embeddings for Generalized Audio-Visual Zero-shot Classification and Retrieval of Videos

March 2, 2020/WACV 2020, Snowmass Village, CO USA

We present an audio-visual multimodal approach for the task of zero-shot learning (ZSL) for classification and retrieval of videos. ZSL has been studied extensively in the recent past but has primarily been limited to visual modality and to images. We demonstrate that both audio and visual modalities

DAVID: Dual-Attentional Video Deblurring

March 2, 2020/WACV 2020, Snowmass Village, CO USA

Blind video deblurring restores sharp frames from a blurry sequence without any prior. It is a challenging task because the blur due to camera shake, object movement and defocusing is heterogeneous in both temporal and spatial dimensions. Traditional methods train on datasets synthesized with a single

Unsupervised and Semi-Supervised Domain Adaptation for Action Recognition from Drones

March 2, 2020/WACV 2020, Snowmass Village, CO USA

We address the problem of human action classification in drone videos. Due to the high cost of capturing and labeling large-scale drone videos with diverse actions, we present unsupervised and semi-supervised domain adaptation approaches that leverage both the existing fully annotated action recognition

Video Person Re-Identification using Learned Clip Similarity Aggregation

March 2, 2020/WACV 2020, Snowmass Village, CO USA

We address the challenging task of video-based person re-identification. Recent works have shown that splitting the video sequences into clips and then aggregating clip-based similarity is appropriate for the task. We show that using a learned clip similarity aggregation function allows filtering out

Asymmetrically Hierarchical Networks with Attentive Interactions for Interpretable Review-based Recommendation

February 12, 2020/The Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI 2020)

Recently, recommender systems have been able to emit substantially improved recommendations by leveraging user-provided reviews. Existing methods typically merge all reviews of a given user (item) into a long document, and then process user and item documents in the same manner. In practice, however,

Deep Unsupervised Binary Coding Networks for Multivariate Time Series Retrieval

February 12, 2020/The Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI 2020)

Multivariate time series data are becoming increasingly ubiquitous in varies real-world applications such as smart city, power plant monitoring, wearable devices, etc. Given the current time series segment, how to retrieve similar segments within the historical data in an efficient and effective manner

Tensorized LSTM with Adaptive Shared Memory for Learning Trends in Multivariate Time Series

February 12, 2020/The Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI 2020)

The problem of learning and forecasting underlying trends in time series data arises in a variety of applications, such as traffic management, energy optimization, etc. In literature, a trend in time series is characterized by the slope and duration, and its prediction is then to forecast the two values

Adversarial Learning of Privacy-Preserving and Task-Oriented Representations

February 7, 2020/AAAI 2020, New York, New York USA

Data privacy has emerged as an important issue as data-driven deep learning has been an essential component of modern machine learning systems. For instance, there could be a potential privacy risk of machine learning systems via the model inversion attack, whose goal is to reconstruct the input data

Interpretable Click-Through Rate Prediction through Hierarchical Attention

February 7, 2020/The 13th ACM International Conference on Web Search and Data Mining (WSDM 2020)

Click-through rate (CTR) prediction is a critical task in online advertising and marketing. For this problem, existing approaches, with shallow or deep architectures, have three major drawbacks. First, they typically lack persuasive rationales to explain the outcomes of the models. Unexplainable predictions

Temporal Context-aware Representation Learning for Question Routing

February 7, 2020/The 13th ACM International Conference on Web Search and Data Mining (WSDM 2020)

Question routing (QR) aims at recommending newly posted questions to the potential answerers who are most likely to answer the questions. The existing approaches that learn users’ expertise from their past question-answering activities usually suffer from challenges in two aspects: 1) multi-faceted expertise

First Field Trial of Distributed Fiber Optical Sensing and High-Speed Communication Over an Operational Telecom Network

January 1, 2020/IEEE Journal of Lightwave Technology

To the best of our knowledge, we present the first field trial of distributed fiber optical sensing (DFOS) and high-speed communication, comprising a coexisting system, over an operation telecom network. Using probabilistic-shaped (PS) DP-144QAM, a 36.8 Tb/s with an 8.28-b/s/Hz spectral efficiency (SE)

Coordination of PV Smart Inverters Using Deep Reinforcement Learning for Grid Voltage Regulation

December 16, 2019/18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019, Boca Raton, FL USA

Increasing adoption of solar photovoltaic (PV) presents new challenges to modern power grid due to its variable and intermittent nature. Fluctuating outputs from PV generation can cause the grid violating voltage operation limits. PV smart inverters (SIs) provide a fast-response method to regulate voltage

Detection of False Data Injection Attacks in Cyber-Physical Systems using Dynamic Invariants

December 16, 2019/18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019, Boca Raton, FL USA

Modern cyber-physical systems are increasingly complex and vulnerable to attacks like false data injection aimed at destabilizing and confusing the systems. We develop and evaluate an attack-detection framework aimed at learning a dynamic invariant network, data-driven temporal causal relationships between

Progressive Processing of System-Behavioral Query

December 13, 2019/The 35th Annual Computer Security Applications Conference (ACSAC 2019)

System monitoring has recently emerged as an effective way to analyze and counter advanced cyber attacks. The monitoring data records a series of system events and provides a global view of system behaviors in an organization. Querying such data to identify potential system risks and malicious behaviors

Contextual Grounding of Natural Language Entities in Images

December 13, 2019/NeurIPS 2019 workshop on Visually Grounded Interaction and Language (ViGIL)

In this paper, we introduce a contextual grounding approach that captures the context in corresponding text entities and image regions to improve the grounding accuracy. Specifically, the proposed architecture accepts pre-trained text token embeddings and image object features from an off-the-shelf object

Multivariate Long-Term State Forecasting in Cyber-Physical Systems: A Sequence to Sequence Approach

December 9, 2019/IEEE Big Data 2019, Los Angeles, CA, USA

Cyber-physical systems (CPS) are ubiquitous in several critical infrastructure applications. Forecasting the state of CPS, is essential for better planning, resource allocation and minimizing operational costs. It is imperative to forecast the state of a CPS multiple steps into the future to afford enough

Adaptive Neural Network for Node Classification in Dynamic Networks

November 11, 2019/The 19th IEEE International Conference on Data Mining (ICDM 2019)

Given a network with the labels for a subset of nodes, transductive node classification targets to predict the labels for the remaining nodes in the network. This technique has been used in a variety of applications such as voxel functionality detection in brain network and group label prediction in

Learning Robust Representations with Graph Denoising Policy Network

November 11, 2019/The 19th IEEE International Conference on Data Mining (ICDM 2019)

Existing representation learning methods based on graph neural networks and their variants rely on the aggregation of neighborhood information, which makes it sensitive to noises in the graph, e.g. erroneous links between nodes, incorrect/missing node features. In this paper, we propose Graph Denoising

Self-Attentive Attributed Network Embedding Through Adversarial Learning

November 11, 2019/The 19th IEEE International Conference on Data Mining (ICDM 2019)

Network embedding aims to learn the low-dimensional representations/embeddings of vertices which preserve the structure and inherent properties of the networks. The resultant embeddings are beneficial to downstream tasks such as vertex classification and link prediction. A vast majority of real-world

Contextual Grounding of Natural Language Phrases in Images

November 5, 2019/arXiv

In this paper, we introduce a contextual grounding approach that captures the context in corresponding text entities and image regions to improve the grounding accuracy. Specifically, the proposed architecture accepts pre-trained text token embeddings and image object features from an off-the-shelf object

On Novel Object Recognition: A Unified Framework for Discriminability and Adaptability

November 4, 2019/CIKM 2019

The rich and accessible labeled data fueled the revolutionary successes of deep learning in object recognition. However, recognizing objects of novel classes with limited supervision information provided, i.e., Novel Object Recognition (NOR), remains a challenging task. We identify in this paper two

Degeneracy in Self-Calibration Revisited and a Deep Learning Solution for Uncalibrated SLAM

November 3, 2019/IROS 2019, The Venetian Macao, Macau, China

Self-calibration of camera intrinsics and radial distortion has a long history of research in the computer vision community. However, it remains rare to see real applications of such techniques to modern Simultaneous Localization And Mapping (SLAM) systems, especially in driving scenarios. In this paper,

Learning 2D to 3D Lifting for Object Detection in 3D for Autonomous Vehicles

November 3, 2019/IROS 2019, The Venetian Macao, Macau, China

We address the problem of 3D object detection from 2D monocular images in autonomous driving scenarios. We propose to lift the 2D images to 3D representations using learned neural networks and leverage existing networks working directly on 3D data to perform 3D object detection and localization. We show

Rethinking Zero-Shot Learning: A Conditional Visual Classification Perspective

October 27, 2019/ICCV 2019

Zero-shot learning (ZSL) aims to recognize instances of unseen classes solely based on the semantic descriptions of the classes. Existing algorithms usually formulate it as a semantic-visual correspondence problem, by learning mappings from one feature space to the other. Despite being reasonable, previous

Domain Adaptation for Structured Output via Discriminative Patch Representations

October 27, 2019/ICCV 2019 - International Conference on Computer Vision 2019, Seoul, Korea

Predicting structured outputs such as semantic segmentation relies on expensive per-pixel annotations to learn supervised models like convolutional neural networks. However, models trained on one data domain may not generalize well to other domains without annotations for model finetuning. To avoid the

GLoSH: Global-Local Spherical Harmonics for Intrinsic Image Decomposition

October 27, 2019/ICCV 2019 - International Conference on Computer Vision 2019, Seoul, Korea

Traditional intrinsic image decomposition focuses on decomposing images into reflectance and shading, leaving surfaces normals and lighting entangled in shading. In this work, we propose a Global-Local Spherical Harmonics (GLoSH) lighting model to improve the lighting component, and jointly predict reflectance

VeCharge: Intelligent Energy Management for Electric Vehicle charging

October 1, 2019/Charge Expo, San Diego, CA USA

2018’s 1.2 million North American charging ports will grow ten times to over 12.6 million by 2027, according to Navigant, which could overwhelm the nation’s grids. DC Fast charging requires grid upgrade to supply the new charging demand. However, since the utilization ratio of those charging station

Wavelength Modulation Spectroscopy Enhanced by Machine Learning for Early Fire Detection

September 30, 2019/IEEE IPC 2019

We proposed and demonstrated a new machine learning algorithm for wavelength modulation spectroscopy to enhance the accuracy of fire detection. The result shows more than 8% of accuracy improvement by analyzing CO/CO 2 2f signals.

Data-Driven Day-Ahead PV Estimation Using Hybrid Deep Learning

September 23, 2019/54th IEEE Industry Applications Society Annual Meeting, Baltimore, MD

Ongoing smart grid activities and associated automation resulted in rich set of data. These data can be utilized for monitoring and estimation of real time photovoltaic (PV) generation. Inherent variability in PV and related impact on power systems is a challenging problem. Improving the accuracy of

Beam Training Optimization in Millimeter-wave Systems under Beamwidth, Modulation and Coding Constraints

September 9, 2019/IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC 2019)

Millimeter-wave (mmWave) bands have the potential to enable significantly high data rates in wireless systems. In order to overcome intense path loss and severe shadowing in these bands, it is essential to employ directional beams for data transmission. Furthermore, it is known that the mmWave channel

Opportunistic Temporal Fair Mode Selection and User Scheduling for Full-duplex Systems

September 9, 2019/IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC 2019)

In-band full-duplex (FD) communications – enabled by recent advances in antenna and RF circuit design – has emerged as one of the promising techniques to improve data rates in wireless systems. One of the major roadblocks in enabling high data rates in FD systems is the inter-user interference (IUI)

A Query System for Efficiently Investigating Complex Attack Behaviors for Enterprise Security

August 30, 2019/45th International Conference on Very Large Data Bases (VLDB 2019)

The need for countering Advanced Persistent Threat (APT) attacks has led to the solutions that ubiquitously monitor system activities in each enterprise host, and perform timely attack investigation over the monitoring data for uncovering the attack sequence. However, existing general-purpose query systems

Model transfer of QoT prediction in optical networks based on artificial neural networks

August 27, 2019/Journal of Optical Communications and Networking

An artificial neural network (ANN) based transfer learning model is built for quality of transmission (QoT) prediction in optical systems feasible with different modulation formats. Knowledge learned from one optical system can be transferred to a similar optical system by adjusting weights in ANN hidden

Heterogeneous Graph Matching Networks for Unknown Malware Detection

August 16, 2019/The 28th International Joint Conference on Artificial Intelligence (IJCAI 2019)

Information systems have widely been the target of malware attacks. Traditional signature-based malicious program detection algorithms can only detect known malware and are prone to evasion techniques such as binary obfuscation, while behavior-based approaches highly rely on the malware training samples

Spatio-Temporal Attentive RNN for Node Classification in Temporal Attributed Graphs

August 16, 2019/The 28th International Joint Conference on Artificial Intelligence (IJCAI 2019)

Node classification in graph-structured data aims to classify the nodes where labels are only available for a subset of nodes. This problem has attracted considerable research efforts in recent years. In real-world applications, both graph topology and node attributes evolve over time. Existing techniques,

Aggregation of BTM Battery Storages to Provide Ancillary Services in Wholesale Electricity Markets

August 12, 2019/Smart Energy Grid Engineering (SEGE), University of Ontario Institute of Technology, Ontario, Canada

The behind the meter battery energy storage systems (BTM-BESSs) have been deployed widely by indus-trial/commercial buildings to manage electricity transaction with utilities in order to reduce customers’ electricity bills. Commercial BTM battery storages are mainly employed to cut the customers’ monthly

Conditional GAN with Discriminative Filter Generation for Text-to-Video Synthesis

August 10, 2019/IJCAI 2019

Developing conditional generative models for text-to-video synthesis is an extremely challenging yet an important topic of research in machine learning. In this work, we address this problem by introducing Text-Filter conditioning Generative Adversarial Network (TFGAN), a conditional GAN model with a

Learning K-way D-dimensional Discrete Embedding for Hierarchical Data Visualization and Retrieval

August 10, 2019/IJCAI 2019

Traditional embedding approaches associate a real-valued embedding vector with each symbol or data point, which is equivalent to applying a linear transformation to “one-hot” encoding of discrete symbols or data objects. Despite simplicity, these methods generate storage-inefficient representations

A Study on Traffic Flow Monitoring Using Optical Fiber Sensor Technology

August 6, 2019/Japan Society of Traffic Engineers

Traffic conditions of the highway, Ya traffic volume meter CCTV Because it is observed in the spot, such as the discovery of traffic disturbances which deviates from the observation spot it may be delayed. The traffic flow has a problem from the point observations data indirectly order to be estimated,

Deep Supervision with Intermediate Concepts (IEEE)

August 1, 2019/IEEE Transactions on Pattern Analysis and Machine Intelligence

Read Deep Supervision with Intermediate Concepts (IEEE). Recent data-driven approaches to scene interpretation predominantly pose inference as an end-to-end black-box mapping, commonly performed by a Convolutional Neural Network (CNN). However, decades of work on perceptual organization in both human

Pose-variant 3D Facial Attribute Generation

July 23, 2019/arXiv

We address the challenging problem of generating facial attributes using a single image in an unconstrained pose. In contrast to prior works that largely consider generation on 2D near-frontal images, we propose a GAN-based framework to generate attributes directly on a dense 3D representation given

Size and Alignment Independent Classification of the High-order Spatial Modes of a Light Beam Using a Convolutional Neural Network

July 16, 2019/Computing Conference 2019

The higher-order spatial modes of a light beam are receiving significant interest. They can be used to further increase the data speeds of high speed optical communication, and for novel optical sensing modalities. As such, the classification of higher-order spatial modes is ubiquitous. Canonical classification

Field and lab experimental demonstration of nonlinear impairment compensation using neural networks

July 10, 2019/Nature Communications

Fiber nonlinearity is one of the major limitations to the achievable capacity in long distance fiber optic transmission systems. Nonlinear impairments are determined by the signal pattern and the transmission system parameters. Deterministic algorithms based on approximating the nonlinear Schrodinger

Decentralized Transactive Energy Auctions with Bandit Learning

July 8, 2019/2019 IEEE PES Transactive Energy Systems Conference (TESC), University of Minnesota, Minneapolis, MN

The power systems worldwide have been embracing the rapid growth of distributed energy resources. Commonly, distributed energy resources exist in the distribution level, such as electric vehicles, rooftop photovoltaic panels, and home battery systems, which cannot be controlled by a centralized entity

Neural-Network-Based G-OSNR Estimation of Probabilistic-Shaped 144QAM Channels in DWDM Metro Network Field Trial

July 7, 2019/OECC/PSC 2019

A two-stage neural network model is applied on captured PS-144QAM raw data to estimate channel G-OSNR in a metro network field trial. We obtained 0.27dB RMSE with first-stage CNN classifier and second-stage ANN regressions.

Energy Predictive Models with Limited Data using Transfer Learning

June 25, 2019/10th ACM International Conference on Future Energy Systems (ACM e-Energy), Phoenix, AZ USA

In this paper, we consider the problem of developing predictive models with limited data for energy assets such as electricity loads, PV power generations, etc. We specifically investigate the cases where the amount of historical data is not sufficient to effectively train the prediction model. We first

Clairvoyant Networks

June 21, 2019/Network Traffic Measurement and Analysis Conference (TMA Conference 2019)

We use the term clairvoyant to refer to networks that provide on-demand visibility for any flow at any time. Traditionally, network visibility is achieved by instrumenting and passively monitoring all flows in a network. SDN networks, by design endowed with full visibility, offer another alternative

A Dataset for High-Level 3D Scene Understanding of Complex Road Scenes in the Top-View

June 17, 2019/Proceedings of CVPR 2019 Workshop on 3D Scene Understanding for Vision, Graphics, and Robotics

We introduce a novel dataset for high-level 3D scene understanding of complex road scenes. Our annotations extend the existing datasets KITTI [5] and nuScenes [1] with semantically and geometrically meaningful attributes like the number of lanes or the existence of, and distance to, intersections, sidewalks

A Parametric Top-View Representation of Complex Road Scenes

June 16, 2019/IEEE Computer Vision and Pattern Recognition (CVPR 2019)

In this paper, we address the problem of inferring the layout of complex road scenes given a single camera as input. To achieve that, we first propose a novel parameterized model of road layouts in a top-view representation, which is not only intuitive for human visualization but also provides an interpretable

Feature Transfer Learning for Face Recognition with Under-Represented Data

June 16, 2019/IEEE Computer Vision and Pattern Recognition (CVPR 2019)

Despite the large volume of face recognition datasets, there is a significant portion of subjects, of which the samples are insufficient and thus under-represented. Ignoring such significant portion results in insufficient training data. Training with under-represented data leads to biased classifiers

Gotta Adapt Em All: Joint Pixel and Feature-Level Domain Adaptation for Recognition in the Wild

June 16, 2019/IEEE Computer Vision and Pattern Recognition (CVPR 2019)

Recent developments in deep domain adaptation have allowed knowledge transfer from a labeled source domain to an unlabeled target domain at the level of intermediate features or input pixels. We propose that advantages may be derived by combining them, in the form of different insights that lead to a

Learning Structure-And-Motion-Aware Rolling Shutter Correction

June 16, 2019/IEEE Computer Vision and Pattern Recognition (CVPR 2019)

An exact method of correcting the rolling shutter (RS) effect requires recovering the underlying geometry, i.e. the scene structures and the camera motions between scanlines or between views. However, the multiple-view geometry for RS cameras is much more complicated than its global shutter (GS) counterpart,

Neural Collaborative Subspace Clustering

June 9, 2019/International Conference on Machine Learning, ICML 2019, Long Beach, CA USA

We introduce the Neural Collaborative Subspace Clustering, a neural model that discovers clusters of data points drawn from a union of low-dimensional subspaces. In contrast to previous attempts, our model runs without the aid of spectral clustering. This makes our algorithm one of the kinds that can

Learning from Rules Performs as Implicit Regularization

June 9, 2019/Thirty-sixth International Conference on Machine Learning (ICML 2019)

In this paper, we study the generalization performance of deep neural networks in learning problems where the given task is governed by a set of rules. We consider two settings of supervised learning and rule-based learning. In supervised learning, the network is trained with pairs of inputs and the

Robust Beam Tracking and Data Communication in Millimeter Wave Mobile Networks

June 3, 2019/The International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt 2019)

Millimeter-wave (mmWave) bands have shown the potential to enable high data rates for next generation mobile networks. In order to cope with high path loss and severe shadowing in mmWave frequencies, it is essential to employ massive antenna arrays and generate narrow transmission patterns (beams). When

Tripping Through Time: Efficient Temporal Localization of Activities in Videos

May 16, 2019/CVPR 2019

Localizing moments in untrimmed videos using language queries is a new task that requires the ability to accurately ground language into video. Existing approaches process the video, often more than once, to localize the activities and are inefficient. In this paper, we present TripNet, an end-to-end

Unsupervised Domain Adaptation for Distance Metric Learning

May 6, 2019/Seventh International Conference on Learning Representations (ICLR 2019)

Unsupervised domain adaptation is a promising avenue to enhance the performance of deep neural networks on a target domain, using labels only from a source domain. However, the two predominant methods, domain discrepancy reduction learning and semi-supervised learning, are not readily applicable when

Learning To Simulate

May 6, 2019/Seventh International Conference on Learning Representations (ICLR 2019)

Simulation is a useful tool in situations where training data for machine learning models is costly to annotate or even hard to acquire. In this work, we propose a reinforcement learning-based method for automatically adjusting the parameters of any (non-differentiable) simulator, thereby controlling

Attentional Heterogeneous Graph Neural Network: Application to Program Reidentification

May 4, 2019/SIAM International Conference on Data Mining (SDM 2019)

Program or process is an integral part of almost every IT/OT system. Can we trust the identity/ID (e.g., executable name) of the program? To avoid detection, malware may disguise itself using the ID of a legitimate program, and a system tool (e.g., PowerShell) used by the attackers may have the fake

Deep Co-Clustering

May 4, 2019/SIAM International Conference on Data Mining (SDM 2019)

Co-clustering partitions instances and features simultaneously by leveraging the duality between them, and it often yields impressive performance improvement over traditional clustering algorithms. The recent development in learning deep representations has demonstrated the advantage in extracting effective

A Deep Spatio-Temporal Fuzzy Neural Network for Passenger Demand Prediction

May 2, 2019/SDM 2019

In spite of its importance, passenger demand prediction is a highly challenging problem, because the demand is simultaneously influenced by the complex interactions among many spatial and temporal factors and other external factors such as weather. To address this problem, we propose a Spatio-TEmporal

Coupled-Core Fiber Design For Enhancing Nonlinearity Tolerance

April 8, 2019/SubOptic 2019

Fiber nonlinearity is a major limitation on the achievable maximum capacity per fiber core. Digital signal processing (DSP) can be used directly to compensate nonlinear impairments, however with limited effectiveness. It is well known that fibers with higher chromatic dispersion (CD) reduce nonlinear

Fiber Nonlinearity Compensation by Neural Networks

April 8, 2019/SubOptic 2019

Neuron network (NN) is proposed to work together with perturbation-based nonlinearity compensation (NLC) algorithm by feeding with intra-channel cross-phase modulation (IXPM) and intra-channel four-wave mixing (IFWM) triplets. Without prior knowledge of the transmission link and signal pulse shaping/baudrate,

Spectrally-Efficient 200G Probabilistically-Shaped 16QAM over 9000km Straight Line Transmission with Flexible Multiplexing Scheme

April 8, 2019/SubOptic 2019

Flexible wavelength-multiplexing technique in backbone submarine networks has been deployed to accommodate the trend of variable-rate modulation formats. In this paper, we propose a new design of flexible-rate transponders in the scenario of flexible multiplexing scheme to achieve near-Shannon performance.

PoLPer: Process-Aware Restriction of Over-Privileged Setuid Calls in Legacy Applications

March 27, 2019/9th ACM Conference on Data and Application Security and Privacy (CODASPY 2019)

Setuid system calls enable critical functions such as user authentications and modular privileged components. Such operations must only be executed after careful validation. However, current systems do not perform rigorous checks, allowing exploitation of privileges through memory corruption vulnerabilities

First Field Trial of Sensing Vehicle Speed, Density, and Road Conditions by Using Fiber Carrying High Speed Data

March 3, 2019/OFC 2019

For the first time, we demonstrate detection of vehicle speed, density, and road conditions using deployed fiber carrying high-speed data transmission, and prove carriers’ large-scale fiber infrastructures can also be used as ubiquitous sensing networks.

On the Performance Metric and Design of Non-Uniformly Shaped Constellation

March 3, 2019/OFC 2019

Asymmetric information is shown to be more accurate in characterizing the performance of quadrant folding shaped (QFS) M-QAM. The performance difference of QFS M-QAM schemes strongly depends on the FEC coding rate, and the optimum FEC coding rate is found to be around ?0.8, which is independent of QFS

Countering Malicious Processes with Process-DNS Association

February 27, 2019/The 26th Annual Network and Distributed System Security Symposium (NDSS 2019)

Modern malware and cyber attacks depend heavily on DNS services to make their campaigns reliable and difficult to track. Monitoring network DNS activities and blocking suspicious domains have been proven an effective technique in countering such attacks. However, recent successful campaigns reveal that

NODOZE: Combatting Threat Alert Fatigue with Automated Provenance Triage

February 27, 2019/The 26th Annual Network and Distributed System Security Symposium (NDSS 2019)

Large enterprises are increasingly relying on threat detection softwares (e.g., Intrusion Detection Systems) to allow them to spot suspicious activities. These softwares generate alerts which must be investigated by cyber analysts to figure out if they are true attacks. Unfortunately, in practice, there

TrackIO: Tracking First Responders Inside-Out

February 26, 2019/16th USENIX Symposium on Networked Systems Design and Implementation (NSDI 2019)

First responders, a critical lifeline of any society, often find themselves in precarious situations. The ability to track them in real-time in unknown indoor environments would significantly contribute to the success of their mission as well as their safety. In this work, we present the design, implementation

Multi-parameter distributed fiber sensing with higherorder optical and acoustic modes

February 18, 2019/Optics Letters

We propose a novel multi-parameter sensing technique based on a Brillouin optical time domain reflectometry in the elliptical-core few-mode fiber, using higher-order optical and acoustic modes. Multiple Brillouin peaks are observed for the backscattering of both the LP01 mode and LP11 mode. We characterize

Transactive Energy Management with Blockchain Smart Contracts for P2P Multi-Settlement Markets

February 18, 2019/IEEE PES 10th Conference on Innovative Smart Grid Technologies (ISGT 2019), Washington D.C. USA

Integration of renewables and energy storage, leading to rise of prosumers, has created localized bidirectional flows. As the result, the utility demand has decreased and traditional centralized controller can no longer realize the optimal performance of ever growing distribution systems. To achieve

Visual Entailment: A Novel Task for Fine-Grained Image Understanding

February 15, 2019/arXiv

Existing visual reasoning datasets, such as Visual Question Answering (VQA), often suffer from biases conditioned on the question, image or answer distributions. The recently proposed CLEVR dataset addresses these limitations and requires fine-grained reasoning, but the dataset is synthetic and consists

A Deep Neural Network for Unsupervised Anomaly Detection and Diagnosis in Multivariate Time Series Data

February 1, 2019/The Thirty-Third AAAI Conference on Artificial Intelligence (AAAI 2019)

Nowadays, multivariate time series data are increasingly collected in various real-world systems, e.g., power plants, wearable devices, etc. Anomaly detection and diagnosis in multivariate time series refer to identifying abnormal status in certain time steps and pinpointing the root causes. Building

Coherent optical wireless communication link employing orbital angular momentum multiplexing in a ballistic and diffusive scattering medium

January 31, 2019/Optics Letters

We experimentally investigate the scattering effect on an 80 Gbit/s orbital angular momentum (OAM) multiplexed optical wireless communication link. The power loss, mode purity, cross talk, and bit error rate performance are measured and analyzed for different OAM modes under scattering levels from ballistic

Attentive Conditional Channel-Recurrent Autoencoding for Attribute-Conditioned Face Synthesis

January 8, 2019/Winter Conference on Applications of Computer Vision (WACV) 2019, Waikoloa Village, Hawaii USA

Attribute-conditioned face synthesis has many potential use cases, such as to aid the identification of a suspect or a missing person. Building on top of a conditional version of VAE-GAN, we augment the pathways connecting the latent space with channel-recurrent architecture, in order to provide not

Memory Warps for Long-Term Online Video Representations and Anticipation

January 8, 2019/Winter Conference on Applications of Computer Vision (WACV) 2019, Waikoloa Village, Hawaii USA

We propose a novel memory-based online video representation that is efficient, accurate and predictive. This is in contrast to prior works that often rely on computationally heavy 3D convolutions, ignore motion when aligning features over time, or operate in an off-line mode to utilize future frames.

41.5-Tb/s Transmission Over 549 km of Field Deployed Fiber Using Throughput Optimized Probabilistic-Shaped 144QAM

January 1, 2019/Journal of Lightwave Technology

We demonstrate high spectral efficiency transmission over 549 km of field-deployed single-mode fiber using probabilistic-shaped 144QAM. We achieved 41.5 Tb/s over the C-band at a spectral efficiency of 9.02 b/s/Hz using 32-Gbaud channels at a channel spacing of 33.33 GHz, and 38.1 Tb/s at a spectral

Battery Degradation Temporal Modeling Using LSTM Networks

December 17, 2018/ICMLA 2018, Orlando, FL USA

Accurate modeling of battery capacity degradation is an important component for both battery manufacturers and energy management systems. In this paper, we develop a battery degradation model using deep learning algorithms. The model is trained with the real data collected from battery storage solutions

Conditioning Neural Networks: A Case Study of Electrical Load Forecasting

December 10, 2018/IEEE Big Data 2018, Seattle, WA, USA

Machine learning tasks typically involve minimizing a loss function that measures the distance of the model output and the ground-truth. In some applications, in addition to the usual loss function, the output must also satisfy certain requirements for further processing. We call such requirements model

Visual Entailment Task for Visually-Grounded Language Learning

December 7, 2018/NeurIPS 2018 workshop on Visually Grounded Interaction and Language (ViGIL)

We introduce a new inference task – Visual Entailment (VE) – which differs from traditional Textual Entailment (TE) tasks whereby a premise is defined by an image, rather than a natural language sentence as in TE tasks. A novel dataset SNLI-VE is proposed for VE tasks based on the Stanford Natural Language

SkyRAN: A Self-Organizing LTE RAN in the Sky

December 4, 2018/The 14th International Conference on emerging Networking EXperiments and Technologies (ACM CoNEXT 2018)

We envision a flexible, dynamic airborne LTE infrastructure built upon Unmanned Autonomous Vehicles (UAVs) that will provide on-demand, on-time, network access, anywhere. In this paper, we design, implement and evaluate SkyRAN, a self-organizing UAV-based LTE RAN (Radio Access Network) that is a key

Optimal Transport Classifier: Defending Against Adversarial Attacks by Regularized Deep Embedding

December 3, 2018/arXiv

Recent studies have demonstrated the vulnerability of deep convolutional neural networks against adversarial examples. Inspired by the observation that the intrinsic dimension of image data is much smaller than its pixel space dimension and the vulnerability of neural networks grows with the input dimension,

Scalable Deep k-Subspace Clustering

December 2, 2018/ACCV 2018, Perth, Australia

Subspace clustering algorithms are notorious for their scalability issues because building and processing large affinity matrices are demanding. In this paper, we introduce a method that simultaneously learns an embedding space along subspaces within it to minimize a notion of reconstruction error, thus

Unseen Object Segmentation in Videos via Transferable Representations

December 2, 2018/ACCV 2018

In order to learn object segmentation models in videos, conventional methods require a large amount of pixel-wise ground truth annotations. However, collecting such supervised data is time-consuming and labor-intensive. In this paper, we exploit existing annotations in source images and transfer such

Leveraging Knowledge Bases for Future Prediction with Memory Comparison Networks

November 15, 2018/AI Communications

Making predictions about what might happen in the future is important for reacting adequately in many situations. For example, observing that Man kidnaps girl may have the consequence that Man kills girl. While this is part of common sense reasoning for humans, it is not obvious how machines

Learning Context-Sensitive Convolutional Filters for Text Processing

October 31, 2018/EMNLP 2018

Convolutional neural networks (CNNs) have recently emerged as a popular building block for natural language processing (NLP). Despite their success, most existing CNN models employed in NLP share the same learned (and static) set of filters for all input sentences. In this paper, we consider an approach

Team Papelo: Transformer Networks at FEVER

October 31, 2018/EMNLP 2018

We develop a system for the FEVER fact extraction and verification challenge that uses a high precision entailment classifier based on transformer networks pretrained with language modeling, to classify a broad set of potential evidence. The precision of the entailment classifier allows us to enhance

Teaching Syntax by Adversarial Distraction

October 31, 2018/EMNLP 2018

Existing entailment datasets mainly pose problems which can be answered without attention to grammar or word order. Learning syntax requires comparing examples where different grammar and word order change the desired classification. We introduce several datasets based on synthetic transformations of

SkyCore: Moving Core to the Edge for Untethered and Reliable UAV-based LTE Networks

October 29, 2018/**BEST PAPER AWARD** The 24th Annual International confrence on Mobile Computing and Networking (MobiCom 2018)

The advances in unmanned aerial vehicle (UAV) technology have empowered mobile operators to deploy LTE base stations (BSs) on UAVs, and provide on-demand, adaptive connectivity to hotspot venues as well as emergency scenarios. However, today’s evolved packet core (EPC) that orchestrates the LTE RAN faces

Demand Charge and Response with Energy Storage

October 29, 2018/IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm) 2018, Aalborg, Denmark

Commercial and industry (C& I) customers incur two types of electricity charges on their bills: one for the amount of energy usage and another one for the maximum demand during certain billing periods. The second charge type is known as Demand Charge (DC), which could account for over half of a customers’

Behavior-based Community Detection: Application to Host Assessment in Enterprise Information Networks

October 26, 2018/Proceedings of the 27th ACM International Conference on Information and Knowledge Management (CIKM 2018)

Behavior-based Community Detection: Application to Host Assessment in Enterprise Information Networks Community detection in complex networks is a fundamental problem that attracts much attention across various disciplines. Previous studies have been mostly focusing on external connections between nodes

Collaborative Alert Ranking for Anomaly Detection

October 26, 2018/Proceedings of the 27th ACM International Conference on Information and Knowledge Management (CIKM 2018)

Given a large number of low-quality heterogeneous categorical alerts collected from an anomaly detection system, how to characterize the complex relationships between different alerts and deliver trustworthy rankings to end users? While existing techniques focus on either mining alert patterns or filtering

TGNet: Learning to Rank Nodes in Temporal Graphs

October 26, 2018/Proceedings of the 27th ACM International Conference on Information and Knowledge Management (CIKM 2018)

Node ranking in temporal networks are often impacted by heterogeneous context from node content, temporal, and structural dimensions. This paper introduces TGNet , a deep-learning framework for node ranking in heterogeneous temporal graphs. TGNet utilizes a variant of Recurrent Neural Network to adapt

NodeMerge: Template Based Efficient Data Reduction For Big-Data Causality Analysis

October 19, 2018/Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security (ACM CCS 2018)

Today’s enterprises are exposed to sophisticated attacks, such as Advanced Persistent Threats~(APT) attacks, which usually consist of stealthy multiple steps. To counter these attacks, enterprises often rely on causality analysis on the system activity data collected from a ubiquitous system monitoring

Learning Gibbs-Regularized Pushforward Density Estimators with a Symmetric KL Objective

October 11, 2018/BayLearn Symposium 2018, Menlo Park, CA USA

We claim that there is currently no satisfactory way to regularize a generative adversarial network (GAN): neither the generator nor discriminator is particularly amenable to the imposition of inductive biases derived from domain knowledge. A generator is effectively a causal model of generationone

Unsupervised Cross Domain Distance Metric Adaptation with Feature Transfer Network

October 11, 2018/BayLearn Symposium 2018, Menlo Park, CA USA

Unsupervised domain adaptation is an attractive avenue to enhance the performance of deep neural networks in a target domain, using labels only from a source domain. However, two predominant methods along this line, namely, domain divergence reduction learning and semi-supervised learning, are not readily

ELI: Empowering LTE with Interference Awareness in Unlicensed Spectrum

September 24, 2018/The 26th IEEE International Conference on Network Protocols (ICNP 2018)

The advent of LTE into the unlicensed spectrum has necessitated the understanding of its operational efficiency when sharing spectrum with different radio access technologies. Our study reveals that LTE, owing to its inherent transmission characteristics, suffers significant performance degradation in

Distributed Temperature and Strain Sensing Using Brillouin Optical Time Domain Reflectometry Over a Few Mode Elliptical Core Optical Fiber

September 24, 2018/OFS-26

We propose a single-ended Brillouin-based sensor in elliptical-core few-mode optical fiber for multi-parameter measurement using spontaneous Brillouin scattering. Distributed sensing of temperature and strain is demonstrated over 0.5 km elliptical-core few-mode fiber.

Intelligent Filtering-Penalty Monitoring and Mitigation for Cascaded WSSs using Ensemble Learning Algorithm

September 23, 2018/ECOC 2018

An ensemble learning algorithm is applied to enhance filtering tolerance of cascaded WSSs in open ROADM environment to demonstrate ~0.8dB Q-factor improvement over MLSE after transmitting over 3200km with 16 ROADMs.

Neuron-Network-based Nonlinearity Compensation Algorithm

September 23, 2018/ECOC 2018

A simplified, system-agnostic NLC algorithm based on a neuron network is proposed to pre-distort symbols at transmitter side to demonstrate ~0.6dB Q improvement after 2800km SMF transmission using 32Gbaud DP-16QAM.

Optimization of Probabilistic Shaping Enabled Transceivers with Large Constellation Sizes for High Capacity Transmission

September 23, 2018/ECOC 2018

We study digital signal processing techniques to optimize the back-to-back performance of large probabilistic shaped constellations. We cover joint optimization of LDPC and constellation shaping, CD pre-compensation, clipping and I/Q imbalance compensation.

Parametric t-Distributed Stochastic Exemplar-centered Embedding

September 10, 2018/ECML 2018

Parametric embedding methods such as parametric t-distributed Stochastic Neighbor Embedding (pt-SNE) enables out-of-sample data visualization without further computationally expensive optimization or approximation. However, pt-SNE favors small mini-batches to train a deep neural network but large mini-batches

Hierarchical Metric Learning and Matching for 2D and 3D Geometric Correspondences

September 8, 2018/European Conference on Computer Vision - ECCV 2018, Munich, Germany

Interest point descriptors have fueled progress on almost every problem in computer vision. Recent advances in deep neural networks have enabled task-specific learned descriptors that outperform hand-crafted descriptors on many problems. We demonstrate that commonly used metric learning approaches do

R2P2: A Reparameterized Pushforward Policy for Diverse, Precise Generative Path Forecasting

September 8, 2018/European Conference on Computer Vision - ECCV 2018, Munich, Germany

We propose a method to forecast a vehicle’s ego-motion as a distribution over spatiotemporal paths, conditioned on features (e.g., from LIDAR and images) embedded in an overhead map. The method learns a policy inducing a distribution over simulated trajectories that is both diverse (produces most paths

Learning to Look around Objects for Top-View Representations of Outdoor Scenes

September 8, 2018/European Conference on Computer Vision – ECCV 2018, Munich, Germany

Given a single RGB image of a complex outdoor road scene in the perspective view, we address the novel problem of estimating an occlusion-reasoned semantic scene layout in the top-view. This challenging problem not only requires an accurate understanding of both the 3D geometry and the semantics of the

Zero-Shot Object Detection

September 8, 2018/European Conference on Computer Vision - ECCV 2018, Munich, Germany

We introduce and tackle the problem of zero-shot object detection (ZSD), which aims to detect object classes which are not observed during training. We work with a challenging set of object classes, not restricting ourselves to similar and/or fine-grained categories as in prior works on zero-shot classification.

Deep Learning IP Network Representations

August 24, 2018/Big-DAMA 2018 - ACM SIGCOMM 2018 Workshop on Big Data Analytics and Machine Learning for Data Communication Networks

We present DIP, a deep learning-based framework to learn structural properties of the Internet, such as node clustering or distance between nodes. Existing embedding-based approaches use linear algorithms on a single source of data, such as latency or hop count information, to approximate the position

DeepConf: Automating Data Center Network Topologies Management with Machine Learning

August 24, 2018/ACM SIGCOMM 2018 Workshop on Network Meets AI & ML (NetAI 2018)

In recent years, many techniques have been developed to improve the performance and efficiency of data center networks. While these techniques provide high accuracy, they are often designed using heuristics that leverage domain-specific properties of the workload or hardware.In this vision paper, we

Deep r-th Root Rank Supervised Joint Binary Embedding for Multivariate Time Series Retrieval

August 23, 2018/KDD 2018 - 24th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

Multivariate time series data are becoming increasingly common in numerous real-world applications, e.g., power plant monitoring, health care, wearable devices, automobiles, etc. As a result, multivariate time series retrieval, i.e., given the current multivariate time series segment, how to obtain its

Learning Deep Network Representations with Adversarially Regularized Autoencoders

August 23, 2018/KDD 2018 - 24th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

The problem of network representation learning, also known as network embedding, arises in many machine learning tasks assuming that there exist a small number of variabilities in the vertex representations which can capture the “semantics” of the original network structure. Most existing network embedding

NetWalk: A Flexible Deep Embedding Approach for Anomaly Detection in Dynamic Networks

August 23, 2018/KDD 2018 – 24th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

Massive and dynamic networks arise in many practical applications such as social media, security and public health. Given an evolutionary network, it is crucial to detect structural anomalies, such as vertices and edges whose “behaviors” deviate from underlying majority of the network, in a real-time

TINET: Transferring Knowledge between Invariant Networks

August 23, 2018/KDD 2018 - 24th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

The latent behavior of an information system that can exhibit extreme events, such as system faults or cyber-attacks, is complex. Recently, the invariant network has shown to be a powerful way of characterizing complex system behaviors. Structures and evolutions of the invariance network, in particular,

SAQL: A Stream-based Query System for Real-Time Abnormal System Behavior Detection

August 17, 2018/The 27th USENIX Security Symposium (USENIX Security 2018)

Recently, advanced cyber attacks, which consist of a sequence of steps that involve many vulnerabilities and hosts, compromise the security of many well-protected businesses. This has led to solutions that ubiquitously monitor system activities in each host (big data) as a series of events and search

The Resilience of Hermite- and Laguerre-Gaussian Modes in Turbulence

August 15, 2018/Journal of Lightwave Technology

Vast geographical distances in Africa are a leading cause for the so-called digital divide due to the high cost of installing fiber. Free-space optical (FSO) communications offer a convenient and higher bandwidth alternative to point-to-point radio microwave links, with the possibility of repurposing

Battery Optimal Approach to Demand Charge Reduction in Behind-The-Meter Energy Management Systems

August 5, 2018/IEEE PES General Meeting 2018, Portland, OR USA

Large monthly demand charge of commercial and industrial entities is a major problem for their economical business. Utilizing a battery by behind-the-meter Energy Management Systems (EMS) has been seen as a solution to demand charge reduction. In state-of-the-art approaches, the EMS maintains sufficient

Exploiting Graph Regularized Multi-dimensional Hawkes Processes for Modeling Events with Spatio-temporal Characteristics

July 19, 2018/The 27th International Joint Conference on Artificial Intelligence (IJCAI-18)

Multi-dimensional Hawkes processes (MHP) has been widely used for modeling temporal events. However, when MHP was used for modeling events with spatio-temporal characteristics, the spatial information was often ignored despite its importance. In this paper, we introduce a framework to exploit MHP for

Baseline Needs More Love: On SimpleWord-Embedding-Based Models and Associated Pooling Mechanisms

July 15, 2018/ACL 2018

Many deep learning architectures have been proposed to model the compositionality in text sequences, requiring substantial number of parameters and expensive computations. However, there has not been a rigorous evaluation regarding the added value of sophisticated compositional functions. In this paper,

AIQL: Enabling Efficient Attack Investigation from System Monitoring Data

July 13, 2018/Proceedings of the 2018 USENIX Annual Technical Conference (ATC 18)

The need for countering Advanced Persistent Threat (APT) attacks has led to solutions that ubiquitously monitor system activities in each host and perform timely attack investigation over the monitoring data for analyzing attack provenance. However, existing query systems based on relational databases

Learning K-way D-dimensional Discrete Code For Compact Embedding Representations

July 10, 2018/ICML 2018

Conventional embedding methods directly associate each symbol with a continuous embedding vector, which is equivalent to applying a linear transformation based on a one-hot encoding of the discrete symbols. Despite its simplicity, such approach yields the number of parameters that grows linearly

LogLens: A Real-time Log Analysis System

July 2, 2018/38th IEEE International Conference on Distributed Computing Systems (ICDCS 2018)

Administrators of most user-facing systems depend on periodic log data to get an idea of the health and status of production applications. Logs report information, which is crucial to diagnose the root cause of complex problems. In this paper, we present a real-time log analysis system called LogLens

Spectrally Efficient Submarine Transmission with Flexible WME

July 2, 2018/OECC 2018

By adjusting single shaping factor in the distribution matcher, probabilistic-shaped M-QAM is reviewed to provide both flex-rate and near-Shannon performance at the given flex-grid bandwidth and filling ratio.

Attend and Interact: Higher-Order Object Interactions for Video Understanding

June 18, 2018/CVPR 2018

Human actions often involve complex interactions across several inter-related objects in the scene. However, existing approaches to fine-grained video understanding or visual relationship detection often rely on single object representation or pairwise object relationships. Furthermore, learning interactions

Fast and Accurate Online Video Object Segmentation via Tracking Parts

June 18, 2018/Conference on Computer Vision and Pattern Recognition (CVPR) 2018, Salt Lake City, UT USA

Online video object segmentation is a challenging task as it entails to process the image sequence timely and accurately. To segment a target object through the video, numerous CNN-based methods have been developed by heavily finetuning on the object mask in the first frame, which is time-consuming for

Learning to Adapt Structured Output Space for Semantic Segmentation

June 18, 2018/Conference on Computer Vision and Pattern Recognition (CVPR) 2018, Salt Lake City, UT USA

Convolutional neural network-based approaches for semantic segmentation rely on supervision with pixel-level ground truth, but may not generalize well to unseen image domains. As the labeling process is tedious and labor intensive, developing algorithms that can adapt source ground truth labels to the

Adaptive and Integared PV Output control with Battery Energy Storage

May 22, 2018/IEEE PES ISGT Asia 2018

An adaptive control system for battery integrated PV generation is designed to reduce the fluctuating in PV power production. The core component of the system is a four-layer power control system (PCS) for Battery Energy Storage (BES). BES responds to the power dispatch commands from PCS and charges/discharges

Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection

May 3, 2018/Proceedings of the 6th International Conference on Learning Representations, Vancouver Convention Center (ICLR 2018)

Unsupervised anomaly detection on multi- or high-dimensional data is of great importance in both fundamental machine learning research and industrial applications, for which density estimation lies at the core. Although previous approaches based on dimensionality reduction followed by density estimation

Co-Regularized Deep Multi-Network Embedding

April 27, 2018/Proceedings of the 2018 World Wide Web Conference (WWW 2018)

Network embedding aims to learn a low-dimensional vector representation for each node in the social and information networks, with the constraint to preserve network structures. Most existing methods focus on single network embedding, ignoring the relationship between multiple networks. In many real-world

Optimal Sizing and Operation of Energy Storage for Demand Charge Management and PV Utilization

April 17, 2018/2018 IEEE PES T&D, Denver, CO USA

This paper presents a method to determine optimal energy and power capacity of distributed Energy Storage Systems (ESS) in behind-the-meter applications to maximize local Photovoltaic (PV) utilization or minimize Demand Charge (DC) cost. The problem is solved as a multi-objective optimization model to

Memory Warps for Learning Long-Term Online Video Representations

March 28, 2018/arXiv

This paper proposes a novel memory-based online video representation that is efficient, accurate and predictive. This is in contrast to prior works that often rely on computationally heavy 3D convolutions, ignore actual motion when aligning features over time, or operate in an off-line mode to utilize

Feature Transfer Learning for Deep Face Recognition with Long-Tail Data

March 23, 2018/arXiv

Real-world face recognition datasets exhibit long-tail characteristics, which results in biased classifiers in conventionally-trained deep neural networks, or insufficient data when long-tail classes are ignored. In this paper, we propose to handle long-tail classes in the training of a face recognition

Channel-Recurrent Autoencoding for Image Modeling

March 14, 2018/WACV 2018, Lake Tahoe, Nevada USA

Despite recent successes in synthesizing faces and bedrooms, existing generative models struggle to capture more complex image types (Figure 1), potentially due to the oversimplification of their latent space constructions. To tackle this issue, building on Variational Autoencoders (VAEs), we integrate

41.5 Tb/s Data Transport over 549 km of Field Deployed Fiber Using Throughput Optimized Probabilistic-Shaped 144QAM to Support Metro Network Capacity Demands

March 11, 2018/OFC 2018

41.5-Tb/s over 549 km of deployed SSMF in Verizon’s network is achieved using probabilistic-shaped 144QAM to optimize throughput at ultra-fine granularity. This is the highest C-band only capacity and spectral efficiency in metro field environment.

ANN-Based Transfer Learning for QoT Prediction in Real-Time Mixed Line-Rate Systems

March 11, 2018/OFC 2018

Quality of transmission prediction for real-time mixed line-rate systems is realized using artificial neural network based transfer learning with SDN orchestrating. 0.42 dB accuracy is achieved with a 1000 to 20 reduction in training samples.

Evolution from 8QAM live traffic to PCS 64-QAM with Neural-Network Based Nonlinearity Compensation on 11000 km Open Subsea Cable

March 11, 2018/OFC 2018

We report on the evolution of the longest segment of FASTER cable at 11,017 km, with 8QAM transponders at 4b/s/Hz spectral efficiency (SE) in service. With offline testing, 6 b/s/Hz is further demonstrated using probabilistically shaped 64QAM, and a novel, low complexity nonlinearity compensation technique

Flex-Rate Transmission using Hybrid Probabilistic and Geometric Shaped 32QAM

March 11, 2018/OFC 2018

A novel algorithm to design geometric shaped 32QAM to work with probabilistic shaping is proposed to approach the Shannon limit within ~0.2 dB in SNR. The experimental results show ~0.2 dB SNR advantage over 64Gbaud PAS-64QAM, and flex-rate transmission demonstrates > 500 km reach improvement over 32QAM.

Universal Hybrid Probabilistic-geometric Shaping Based on Two-dimensional Distribution Matchers

March 11, 2018/OFC 2018

We propose universal distribution matchers applicable to any two-dimensional signal constellation. We experimentally demonstrate that the performance of 32-ary QAM, based on hybrid probabilistic-geometric shaping, is superior to probabilistically shaped 32QAM and regular 32QAM.

SVBRDF-Invariant Shape and Reflectance Estimation from a Light-Field Camera

February 22, 2018/IEEE Transactions on Pattern Analysis and Machine Intelligence

Light-field cameras have recently emerged as a powerful tool for one-shot passive 3D shape capture. However, obtaining the shape of glossy objects like metals or plastics remains challenging, since standard Lambertian cues like photo-consistency cannot be easily applied. In this paper, we derive a spatially-varying

Towards a Timely Causality Analysis for Enterprise Security

February 21, 2018/Proceedings of Network and Distributed Systems Security (NDSS) Symposium 2018

The increasingly sophisticated Advanced Persistent Threat (APT) attacks have become a serious challenge for enterprise IT security. Attack causality analysis, which tracks multi-hop causal relationships between files and processes to diagnose attack provenances and consequences, is the first step towards

Constellation Design with Geometric and Probabilistic Shaping

February 15, 2018/Optics Communications

A systematic study, including theory, simulation and experiments, is carried out to review the generalized pairwise optimization algorithm for designing optimized constellation. In order to verify its effectiveness, the algorithm is applied in three testing cases: 2-dimensional 8 quadrature amplitude

Joint Pixel and Feature-level Domain Adaptation in the Wild

February 5, 2018/arXiv

Recent developments in deep domain adaptation have allowed knowledge transfer from a labeled source domain to an unlabeled target domain at the level of intermediate features or input pixels. We propose that advantages may be derived by combining them, in the form of different insights that lead to a

Adaptive Feature Abstraction for Translating Video to Text

February 2, 2018/The Thirty-Second AAAI Conference on Artificial Intelligence

Previous models for video captioning often use the output from a specific layer of a Convolutional Neural Network (CNN) as video features. However, the variable context-dependent semantics in the video may make it more appropriate to adaptively select features from the multiple CNN layers. We propose

Video Generation From Text

February 2, 2018/The Thirty-Second AAAI Conference on Artificial Intelligence

Generating videos from text has proven to be a significant challenge for existing generative models. We tackle this problem by training a conditional generative model to extract both static and dynamic information from text. This is manifested in a hybrid framework, employing a Variational Autoencoder

Adaptive Memory Networks

February 1, 2018/arXiv

Adaptive Memory Networks We present Adaptive Memory Networks (AMN) that processes input-question pairs to dynamically construct a network architecture optimized for lower inference times for Question Answering (QA) tasks. AMN processes the input story to extract entities and stores them in memory banks.

Learning random-walk label propagation for weakly-supervised semantic segmentation

February 1, 2018/arXiv

Large-scale training for semantic segmentation is challenging due to the expense of obtaining training data for this task relative to other vision tasks. We propose a novel training approach to address this difficulty. Given cheaply-obtained sparse image labelings, we propagate the sparse labels to produce

SkyLiTE: End-to-End Design of Low-altitutde UAV Networks for Providing LTE Connectivity

January 19, 2018/arXiv

Un-manned aerial vehicle (UAVs) have the potential to change the landscape of wide-area wireless connectivity by bringing them to areas where connectivity was sparing or non-existent (e.g. rural areas) or has been compromised due to disasters. While Google’s Project Loon and Facebook’s Project Aquila

illiad: InteLLigent Invariant and Anomaly Detection in Cyber-Physical Systems

January 16, 2018/ACM Transactions on Intelligent Systems and Technology (TIST)

Cyber-physical systems (CPSs) are today ubiquitous in urban environments. Such systems now serve as the backbone to numerous critical infrastructure applications, from smart grids to IoT installations. Scalable and seamless operation of such CPSs requires sophisticated tools for monitoring the time series

Design and Comparison of Advanced Modulation Formats Based on Generalized Mutual Information

January 15, 2018/Journal of Lightwave Technology

Generalized mutual information (GMI) has been comprehensively studied in multidimensional constellation and probabilistic-shaped (PS) constellation together with different forward error correction (FEC) coding schemes. The simulation results confirm that GMI is an efficient and accurate tool to compare

A 4D Light-Field Dataset & CNN Architectures for Material Recognition

August 24, 2016/ECCV 2016

We introduce a new light-field dataset of materials and take advantage of the recent success of deep learning to perform material recognition on the 4D light field. Our dataset contains 12 material categories, each with 100 images taken with a Lytro Illum, from which we extract about 30,000 patches in

A Continuous Occlusion Model for Road Scene Understanding

June 27, 2016/2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

We present a physically interpretable 3D model for handling occlusions with applications to road scene understanding. Given object detection and SFM point tracks, our unified model probabilistically assigns point tracks to objects and reasons about object detection scores and bounding boxes. It uniformly

WarpNet: Weakly Supervised Matching for Single-View Reconstruction

June 1, 2016/CVPR 2016

Our WarpNet matches images of objects in fine-grained datasets without using part annotations. It aligns an object in one image with a different object in another by exploiting a fine-grained dataset to create artificial data for training a Siamese network with an unsupervised discriminative learning

Atomic Scenes for Scalable Traffic Scene Recognition in Monocular Videos

March 7, 2016/CVPR 2016

We propose a novel framework for monocular traffic scene recognition, relying on a decomposition into high-order and atomic scenes to meet those challenges. High-order scenes carry semantic meaning useful for AWS applications, while atomic scenes are easy to learn and represent elemental behaviors based

Attribute2Image: Conditional Image Generation From Visual Attributes

December 2, 2015/ECCV 2016, The 14th European Conference on Computer Vision (2016)

This paper investigates a novel problem of generating images from visual attributes. We model the image as a composite of foreground and background and develop a layered generative model with disentangled latent variables that can be learned end-to-end using a variational auto-encoder. We experiment