Zhengzhang Chen NEC Labs America

Zhengzhang Chen

Senior Researcher

Data Science and System Security


Interdependent Causal Networks for Root Cause Localization

Interdependent Causal Networks for Root Cause Localization 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 construction of a single effective isolated causal network, whereas many real-world systems are complex and exhibit interdependent structures (i.e., multiple networks of a system are interconnected by cross-network links). In interdependent networks, the malfunctioning effects of problematic system entities can propagate to other networks or different levels of system entities. Consequently, ignoring the interdependency results in suboptimal root cause analysis outcomes.In this paper, we propose REASON, a novel framework that enables the automatic discovery of both intra-level (i.e., within-network) and inter-level (i.e., across-network) causal relationships for root cause localization. REASON consists of Topological Causal Discovery (TCD) and Individual Causal Discovery (ICD). The TCD component aims to model the fault propagation in order to trace back to the root causes. To achieve this, we propose novel hierarchical graph neural networks to construct interdependent causal networks by modeling both intra-level and inter-level non-linear causal relations. Based on the learned interdependent causal networks, we then leverage random walk with restarts to model the network propagation of a system fault. The ICD component focuses on capturing abrupt change patterns of a single system entity. This component examines the temporal patterns of each entity’s metric data (i.e., time series), and estimates its likelihood of being a root cause based on the Extreme Value theory. Combining the topological and individual causal scores, the top K system entities are identified as root causes. Extensive experiments on three real-world datasets validate the effectiveness of the proposed framework.

Incremental Causal Graph Learning for Online Root Cause Localization

Incremental Causal Graph Learning for Online Root Cause Localization 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 offline RCA algorithms, which often require manually initiating the RCA process, a significant amount of time and data to train a robust model, and then being retrained from scratch for a new system fault.In this paper, we propose CORAL, a novel online RCA framework that can automatically trigger the RCA process and incrementally update the RCA model. CORAL consists of Trigger Point Detection, Incremental Disentangled Causal Graph Learning, and Network Propagation-based Root Cause Localization. The Trigger Point Detection component aims to detect system state transitions automatically and in near-real-time. To achieve this, we develop an online trigger point detection approach based on multivariate singular spectrum analysis and cumulative sum statistics. To efficiently update the RCA model, we propose an incremental disentangled causal graph learning approach to decouple the state-invariant and state-dependent information. After that, CORAL applies a random walk with restarts to the updated causal graph to accurately identify root causes. The online RCA process terminates when the causal graph and the generated root cause list converge. Extensive experiments on three real-world datasets demonstrate the effectiveness and superiority of the proposed framework.

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

Beyond One Model Fits All: A Survey of Domain Specialization for Large Language Models 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 just a “chatbot”, and use it as an assistant or even replacement for domain experts and tools in specific domains such as healthcare, finance, and education. However, directly applying LLMs to solve sophisticated problems in specific domains meets many hurdles, caused by the heterogeneity of domain data, the sophistication of domain knowledge, the uniqueness of domain objectives, and the diversity of the constraints (e.g., various social norms, cultural conformity, religious beliefs, and ethical standards in the domain applications). To fill such a gap, explosively increase research, and practices have been conducted in very recent years on the domain specialization of LLMs, which, however, calls for a comprehensive and systematic review to better summarizes and guide this promising domain. In this survey paper, first, we propose a systematic taxonomy that categorizes the LLM domain specialization techniques based on the accessibility to LLMs and summarizes the framework for all the subcategories as well as their relations and differences to each other. We also present a comprehensive taxonomy of critical application domains that can benefit from specialized LLMs, discussing their practical significance and open challenges. Furthermore, we offer insights into the current research status and future trends in this area.

Towards Robust Graph Neural Networks via Adversarial Contrastive Learning

Towards Robust Graph Neural Networks via Adversarial Contrastive Learning 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 vulnerabilities and defend GNN against the adversarial attacks? Adversarial training has shown to be effective in improving the robustness of traditional Deep Neural Networks (DNNs). However, existing adversarial training works mainly focus on the image data, which consists of continuous features, while the features and structures of graph data are often discrete. Moreover, rather than assuming each sample is independent and identically distributed as in DNN, GNN leverages the contextual information across the graph (e.g., neighborhoods of a node). Thus, existing adversarial training techniques cannot be directly applied to defend GNN. In this paper, we propose ContrastNet, an effective adversarial defense framework for GNN. In particular, we propose an adversarial contrastive learning method to train the GNN over the adversarial space. To further improve the robustness of GNN, we investigate the latent vulnerabilities in every component of a GNN encoder and propose corresponding refining strategies. Extensive experiments on three public datasets demonstrate the effectiveness of ContrastNet in improving the robustness of popular GNN variants, such as Graph Convolutional Network and GraphSage, under various types of adversarial attacks.

Multi-source Inductive Knowledge Graph Transfer

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 (nodes) and their relationships (edges). Transferring knowledge from existing well-curated source graphs can help construct the target graph of newly-deployed systems faster and better which no doubt will benefit downstream tasks such as link prediction and anomaly detection for new systems. However, current graph transferring methods are either based on a single source, which does not sufficiently consider multiple available sources, or not selectively learns from these sources. In this paper, we propose MSGT-GNN, a graph knowledge transfer model for efficient graph link prediction from multiple source graphs. MSGT-GNN consists of two components: the Intra-Graph Encoder, which embeds latent graph features of system entities into vectors, and the graph transferor, which utilizes graph attention mechanism to learn and optimize the embeddings of corresponding entities from multiple source graphs, in both node level and graph level. Experimental results on multiple real-world datasets from various domains show that MSGT-GNN outperforms other baseline approaches in the link prediction and demonstrate the merit of attentive graph knowledge transfer and the effectiveness of MSGT-GNN.

Towards Learning Disentangled Representations for Time Series

Towards Learning Disentangled Representations for Time Series 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 factors can bring semantic-rich representations of time series and further enhance interpretability. However, directly adopting the sequential models, such as Long Short-Term Memory Variational AutoEncoder (LSTM-VAE), would encounter a Kullback?Leibler (KL) vanishing problem: the LSTM decoder often generates sequential data without efficiently using latent representations, and the latent spaces sometimes could even be independent of the observation space. And traditional disentanglement methods may intensify the trend of KL vanishing along with the disentanglement process, because they tend to penalize the mutual information between the latent space and the observations. In this paper, we propose Disentangle Time-Series, a novel disentanglement enhancement framework for time series data. Our framework achieves multi-level disentanglement by covering both individual latent factors and group semantic segments. We propose augmenting the original VAE objective by decomposing the evidence lower-bound and extracting evidence linking factorial representations to disentanglement. Additionally, we introduce a mutual information maximization term between the observation space to the latent space to alleviate the KL vanishing problem while preserving the disentanglement property. Experimental results on five real-world IoT datasets demonstrate that the representations learned by DTS achieve superior performance in various tasks with better interpretability.

Superclass-Conditional Gaussian Mixture Model for Coarse-To-Fine Few-Shot Learning

Superclass-Conditional Gaussian Mixture Model for Coarse-To-Fine Few-Shot Learning 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 necessitates adaptation of a “coarsely” pre-trained model to new tasks with a few “finer-grained” training labels. However, coarsely supervised pre-training tends to suppress intra-class variation, which is vital for cross-granularity adaptation. In this paper, we develop a training framework underlain by a novel superclass-conditional Gaussian mixture model (SCGM). SCGM imitates the generative process of samples from hierarchies of classes through latent variable modeling of the fine-grained subclasses. The framework is agnostic to the encoders and only adds a few distribution related parameters, thus is efficient, and flexible to different domains. The model parameters are learned end-to-end by maximum-likelihood estimation via a principled Expectation-Maximization algorithm. Extensive experiments on benchmark datasets and a real-life medical dataset indicate the effectiveness of our method.

Structural Temporal Graph Neural Networks for Anomaly Detection in Dynamic Graphs

Structural Temporal Graph Neural Networks for Anomaly Detection in Dynamic Graphs 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 related to the target nodes in a given time window. In this paper, we propose StrGNN, an end-to-end structural temporal Graph Neural Network model for detecting anomalous edges in dynamic graphs. In particular, we first extract the h-hop enclosing subgraph centered on the target edge and propose a node labeling function to identify the role of each node in the subgraph. Then, we leverage the graph convolution operation and Sortpooling layer to extract the fixed-size feature from each snapshot/timestamp. Based on the extracted features, we utilize the Gated Recurrent Units to capture the temporal information for anomaly detection. We fully implement StrGNN and deploy it into a real enterprise security system, and it greatly helps detect advanced threats and optimize the incident response. Extensive experiments on six benchmark datasets also demonstrate the effectiveness of StrGNN.

Interpreting Convolutional Sequence Model by Learning Local Prototypes with Adaptation Regularization

Interpreting Convolutional Sequence Model by Learning Local Prototypes with Adaptation Regularization n 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 data modeling, due to the highly recursive functions, the underlying rationales of their predictions are difficult to explain. Thus, in this paper, we aim to develop a sequence modeling approach that explains its own predictions by breaking input sequences down into evidencing segments (i.e., sub-sequences) in its reasoning. To this end, we build our model upon convolutional neural networks, which, in their vanilla forms, associates local receptive fields with outputs in an obscure manner. To unveil it, we resort to case-based reasoning, and design prototype modules whose units (i.e., prototypes) resemble exemplar segments in the problem domain. Each prediction is obtained by combining the comparisons between the prototypes and the segments of an input. To enhance interpretability, we propose a training objective that delicately adapts the distribution of prototypes to the data distribution in latent spaces, and design an algorithm to map prototypes to human-understandable segments. Through extensive experiments in a variety of domains, we demonstrate that our model can achieve high interpretability generally, together with a competitive accuracy to the state-of-the-art approaches.

Convolutional Transformer based Dual Discriminator Generative Adversarial Networks for Video Anomaly Detection

Convolutional Transformer based Dual Discriminator Generative Adversarial Networks for Video Anomaly Detection 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 patterns effectively and efficiently. Moreover, existing works seldom explicitly consider the local consistency at frame level and global coherence of temporal dynamics in video sequences. To this end, we propose Convolutional Transformer based Dual Discriminator Generative Adversarial Networks (CT-D2GAN) to perform unsupervised video anomaly detection. Specifically, we first present a convolutional transformer to perform future frame prediction. It contains three key components, i.e., a convolutional encoder to capture the spatial information of the input video clips, a temporal self-attention module to encode the temporal dynamics, and a convolutional decoder to integrate spatio-temporal features and predict the future frame. Next, a dual discriminator based adversarial training procedure, which jointly considers an image discriminator that can maintain the local consistency at frame-level and a video discriminator that can enforce the global coherence of temporal dynamics, is employed to enhance the future frame prediction. Finally, the prediction error is used to identify abnormal video frames. Thoroughly empirical studies on three public video anomaly detection datasets, i.e., UCSD Ped2, CUHK Avenue, and Shanghai Tech Campus, demonstrate the effectiveness of the proposed adversarial spatio-temporal modeling framework.