Xiang Zhang works at Pennsylvania State University.


Dynamic Causal Discovery in Imitation Learning

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 from two aspects: 1) agents in imitation learning are usually implemented as deep neural networks, which are black-box models and lack interpretability; 2) the latent causal mechanism behind agents’ decisions may vary along the trajectory, rather than staying static throughout time steps. To increase transparency and offer better interpretability of the neural agent, we propose to expose its captured knowledge in the form of a directed acyclic causal graph, with nodes being action and state variables and edges denoting the causal relations behind predictions. Furthermore, we design this causal discovery process to be state-dependent, enabling it to model the dynamics in latent causal graphs. Concretely, we conduct causal discovery from the perspective of Granger causality and propose a self-explainable imitation learning framework, CAIL. The proposed framework is composed of three parts: a dynamic causal discovery module, a causality encoding module, and a prediction module, and is trained in an end-to-end manner. After the model is learned, we can obtain causal relations among states and action variables behind its decisions, exposing policies learned by it. Experimental results on both synthetic and real-world datasets demonstrate the effectiveness of the proposed CAIL in learning the dynamic causal graphs for understanding the decision-making of imitation learning meanwhilemaintaining high prediction accuracy.

Skill Disentanglement for Imitation Learning from Suboptimal Demonstrations

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. Usually, a trade-off between demonstration quality and quantity needs to be made. Targeting this problem, in this work we consider the imitation of sub-optimal demonstrations, with both a small clean demonstration set and a large noisy set. Some pioneering works have been proposed, but they suffer from many limitations, e.g., assuming a demonstration to be of the same optimality throughout time steps and failing to provide any interpretation w.r.t knowledge learned from the noisy set. Addressing these problems, we propose method by evaluating and imitating at the sub-demonstration level, encoding action primitives of varying quality into different skills. Concretely, SDIL consists of a high-level controller to discover skills and a skill-conditioned module to capture action-taking policies and is trained following a two-phase pipeline by first discovering skills with all demonstrations and then adapting the controller to only the clean set. A mutual-information-based regularization and a dynamic sub-demonstration optimality estimator are designed to promote disentanglement in the skill space. Extensive experiments are conducted over two gym environments and a real-world healthcare dataset to demonstrate the superiority of SDIL in learning from sub-optimal demonstrations and its improved interpretability by examining learned skills.

Personalized Federated Learning via Heterogeneous Modular Networks

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 among local clients diverges. To address this issue, we present Federated Modular Network (FedMN), a novel PFL approach that adaptively selects sub-modules from a module pool to assemble heterogeneous neural architectures for different clients. FedMN adopts a light-weighted routing hypernetwork to model the joint distribution on each client and produce the personalized selection of the module blocks for each client. To reduce the communication burden in existing FL, we develop an efficient way to interact between the clients and the server. We conduct extensive experiments on the real-world test beds and the results show both effectiveness and efficiency of the proposed FedMN over the baselines.

Dynamic Causal Discovery in Imitation Learning

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 implemented as deep neural networks (DNNs), which are black-box models and lack in interpretability, 2) the latent causal mechanism behind agents’ decisions may vary along the trajectory, rather than staying static throughout time steps. To address these difficulties, in this paper, we propose a self-explaining imitation framework, which can expose causal relations among states and action variables behind its decisions. Specifically, a dynamic causal discovery module is designed to extract the causal graph basing on historical trajectory and current states at each time step, and a causality encoding module is designed to model the interactions among variables with discovered causal edges. After encoding causality into variable embeddings, a prediction model conducts the imitation learning on top of obtained representations. These three components are trained end-to-end, and discovered causal edges can provide interpretations on rules captured by the agent. Comprehensive experiments are conducted on the simulation dataset to analyze its causal discovery capacity, and we further test it on a real-world medical dataset MIMIC-IV. Experimental results demonstrate its potential of providing explanations behind decisions.

InfoGCL: Information-Aware Graph Contrastive Learning

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 create two contrastive views, they differ in a variety of view augmentations, architectures, and objectives. It remains an open question how to build your graph contrastive learning model from scratch for particular graph tasks and datasets. In this work, we aim to fill this gap by studying how graph information is transformed and transferred during the contrastive learning process, and proposing an information-aware graph contrastive learning framework called InfoGCL. The key to the success of the proposed framework is to follow the Information Bottleneck principle to reduce the mutual information between contrastive parts while keeping task-relevant information intact at both the levels of the individual module and the entire framework so that the information loss during graph representation learning can be minimized. We show for the first time that all recent graph contrastive learning methods can be unified by our framework. Based on theoretical and empirical analysis on benchmark graph datasets, we show that InfoGCL achieves state-of-the-art performance in the settings of both graph classification and node classification tasks.

Deep Multi-Instance Contrastive Learning with Dual Attention for Anomaly Precursor Detection

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 anomaly types is hard. Second, the precise annotation of anomaly incipient periods is lacking. Third, the interpretable tools to diagnose the precursor symptoms are lacking. Despite some recent progresses, few of the existing approaches can jointly resolve these challenges. In this paper, we propose MCDA, a deep multi-instance contrastive learning approach with dual attention, to detect anomaly precursor. MCDA utilizes multi-instance learning to model the uncertainty of precursor period and employs recurrent neural network with tensorized hidden states to extract precursor features encoded in temporal dynamics as well as the correlations between different pairs of time series. A dual attention mechanism on both temporal aspect and time series variables is developed to pinpoint the time period and the sensors the precursor symptoms are involved in. A contrastive loss is designed to address the issue that annotated anomalies are few. To the best of our knowledge, MCDA is the first method studying the problem of ‘when’ and ‘where’ for the anomaly precursor detection simultaneously. Extensive experiments on both synthetic and real datasets demonstrate the effectiveness of MCDA.

Learning to Drop: Robust Graph Neural Network via Topological Denoising

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 graphs are often noisy and contain task-irrelevant edges, which may lead to suboptimal generalization performance in the learned GNN models. In this paper, we propose PTDNet, a parameterized topological denoising network, to improve the robustness and generalization performance of GNNs by learning to drop task-irrelevant edges. PTDNet prunes task-irrelevant edges by penalizing the number of edges in the sparsified graph with parameterized networks. To take into consideration the topology of the entire graph, the nuclear norm regularization is applied to impose the low-rank constraint on the resulting sparsified graph for better generalization. PTDNet can be used as a key component in GNN models to improve their performances on various tasks, such as node classification and link prediction. Experimental studies on both synthetic and benchmark datasets show that PTDNet can improve the performance of GNNs significantly and the performance gain becomes larger for more noisy datasets.

Multi-Task Recurrent Modular Networks

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, which is not flexible enough to model the dynamic task relationships and lacks generalization abilities to novel test-time scenarios. We propose multi-task recurrent modular networks (MT-RMN) that can be incorporated in any multi-task recurrent models to address the above drawbacks. MT-RMN consists of a shared encoder and multiple task-specific decoders, and recurrently operates over time. For better flexibility, it modularizes the encoder into multiple layers of sub-networks and dynamically controls the connection between these sub-networks and the decoders at different time steps, which provides the recurrent networks with varying degrees of parameter sharing for tasks with dynamic relatedness. For the generalization ability, MT-RMN aims to discover a set of generalizable sub-networks in the encoder that are assembled in different ways for different tasks. The policy networks augmented with the differentiable routers are utilized to make the binary connection decisions between the sub-networks. The experimental results on three multi-task sequence processing datasets consistently demonstrate the effectiveness of MT-RMN.

Parameterized Explainer for Graph Neural Network

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 a single instance, e.g. a node or a graph. As a result, the explanation generated is painstakingly customized for each instance. The unique explanation interpreting each instance independently is not sufficient to provide a global understanding of the learned GNN model, leading to the lack of generalizability and hindering it from being used in the inductive setting. Besides, as it is designed for explaining a single instance, it is challenging to explain a set of instances naturally (e.g., graphs of a given class). In this study, we address these key challenges and propose PGExplainer, a parameterized explainer for GNNs. PGExplainer adopts a deep neural network to parameterize the generation process of explanations, which enables PGExplainer a natural approach to explaining multiple instances collectively. Compared to the existing work, PGExplainer has better generalization ability and can be utilized in an inductive setting easily. Experiments on both synthetic and real-life datasets show highly competitive performance with up to 24.7% relative improvement in AUC on explaining graph classification over the leading baseline.

Tensorized LSTM with Adaptive Shared Memory for Learning Trends in Multivariate Time Series

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 of the subsequent trend given historical data of the time series. For this problem, existing approaches mainly deal with the case in univariate time series. However, in many real-world applications, there are multiple variables at play, and handling all of them at the same time is crucial for an accurate prediction. A natural way is to employ multi-task learning (MTL) techniques in which the trend learning of each time series is treated as a task. The key point of MTL is to learn task relatedness to achieve better parameter sharing, which however is challenging in trend prediction task. First, effectively modeling the complex temporal patterns in different tasks is hard as the temporal and spatial dimensions are entangled. Second, the relatedness among tasks may change over time. In this paper, we propose a neural network, DeepTrends, for multivariate time series trend prediction. The core module of DeepTrends is a tensorized LSTM with adaptive shared memory (TLASM). TLASM employs the tensorized LSTM to model the temporal patterns of long-term trend sequences in an MTL setting. With an adaptive shared memory, TLASM is able to learn the relatedness among tasks adaptively, based upon which it can dynamically vary degrees of parameter sharing among tasks. To further consider short-term patterns, DeepTrends utilizes a multi-task 1dCNN to learn the local time series features, and employs a task-specific sub-network to learn a mixture of long-term and short-term patterns for trend prediction. Extensive experiments on real datasets demonstrate the effectiveness of the proposed model.