Collaborating Authors

Zhu, He

Graph Collaborative Reasoning Artificial Intelligence

Graphs can represent relational information among entities and graph structures are widely used in many intelligent tasks such as search, recommendation, and question answering. However, most of the graph-structured data in practice suffers from incompleteness, and thus link prediction becomes an important research problem. Though many models are proposed for link prediction, the following two problems are still less explored: (1) Most methods model each link independently without making use of the rich information from relevant links, and (2) existing models are mostly designed based on associative learning and do not take reasoning into consideration. With these concerns, in this paper, we propose Graph Collaborative Reasoning (GCR), which can use the neighbor link information for relational reasoning on graphs from logical reasoning perspectives. We provide a simple approach to translate a graph structure into logical expressions, so that the link prediction task can be converted into a neural logic reasoning problem. We apply logical constrained neural modules to build the network architecture according to the logical expression and use back propagation to efficiently learn the model parameters, which bridges differentiable learning and symbolic reasoning in a unified architecture. To show the effectiveness of our work, we conduct experiments on graph-related tasks such as link prediction and recommendation based on commonly used benchmark datasets, and our graph collaborative reasoning approach achieves state-of-the-art performance.

ReLACE: Reinforcement Learning Agent for Counterfactual Explanations of Arbitrary Predictive Models Artificial Intelligence

The demand for explainable machine learning (ML) models has been growing rapidly in recent years. Amongst the methods proposed to associate ML model predictions with human-understandable rationale, counterfactual explanations are one of the most popular. They consist of post-hoc rules derived from counterfactual examples (CFs), i.e., modified versions of input samples that result in alternative output responses from the predictive model to be explained. However, existing CF generation strategies either exploit the internals of specific models (e.g., random forests or neural networks), or depend on each sample's neighborhood, which makes them hard to be generalized for more complex models and inefficient for larger datasets. In this work, we aim to overcome these limitations and introduce a model-agnostic algorithm to generate optimal counterfactual explanations. Specifically, we formulate the problem of crafting CFs as a sequential decision-making task and then find the optimal CFs via deep reinforcement learning (DRL) with discrete-continuous hybrid action space. Differently from other techniques, our method is easily applied to any black-box model, as this resembles the environment that the DRL agent interacts with. In addition, we develop an algorithm to extract explainable decision rules from the DRL agent's policy, so as to make the process of generating CFs itself transparent. Extensive experiments conducted on several datasets have shown that our method outperforms existing CF generation baselines.

TENT: Text Classification Based on ENcoding Tree Learning Artificial Intelligence

Text classification is a primary task in natural language processing (NLP). Recently, graph neural networks (GNNs) have developed rapidly and been applied to text classification tasks. Although more complex models tend to achieve better performance, research highly depends on the computing power of the device used. In this article, we propose TENT ( to obtain better text classification performance and reduce the reliance on computing power. Specifically, we first establish a dependency analysis graph for each text and then convert each graph into its corresponding encoding tree. The representation of the entire graph is obtained by updating the representation of the non-leaf nodes in the encoding tree. Experimental results show that our method outperforms other baselines on several datasets while having a simple structure and few parameters.

FakeSafe: Human Level Data Protection by Disinformation Mapping using Cycle-consistent Adversarial Network Artificial Intelligence

The concept of disinformation is to use fake messages to confuse people in order to protect the real information. This strategy can be adapted into data science to protect valuable private and sensitive data. Huge amount of private data are being generated from personal devices such as smart phone and wearable in recent years. Being able to utilize these personal data will bring big opportunities to design personalized products, conduct precision healthcare and many other tasks that were impossible in the past. However, due to privacy, safety and regulation reasons, it is often difficult to transfer or store data in its original form while keeping them safe. Building a secure data transfer and storage infrastructure to preserving privacy is costly in most cases and there is always a concern of data security due to human errors. In this study, we propose a method, named FakeSafe, to provide human level data protection using generative adversarial network with cycle consistency and conducted experiments using both benchmark and real world data sets to illustrate potential applications of FakeSafe.

Robustness to Adversarial Attacks in Learning-Enabled Controllers Artificial Intelligence

Learning-enabled controllers used in cyber-physical systems (CPS) are known to be susceptible to adversarial attacks. Such attacks manifest as perturbations to the states generated by the controller's environment in response to its actions. We consider state perturbations that encompass a wide variety of adversarial attacks and describe an attack scheme for discovering adversarial states. To be useful, these attacks need to be natural, yielding states in which the controller can be reasonably expected to generate a meaningful response. We consider shield-based defenses as a means to improve controller robustness in the face of such perturbations. Our defense strategy allows us to treat the controller and environment as black-boxes with unknown dynamics. We provide a two-stage approach to construct this defense and show its effectiveness through a range of experiments on realistic continuous control domains such as the navigation control-loop of an F16 aircraft and the motion control system of humanoid robots.

ART: Abstraction Refinement-Guided Training for Provably Correct Neural Networks Artificial Intelligence

Artificial neural networks (ANNs) have demonstrated remarkable utility in a variety of challenging machine learning applications. However, their complex architecture makes asserting any formal guarantees about their behavior difficult. Existing approaches to this problem typically consider verification as a post facto white-box process, one that reasons about the safety of an existing network through exploration of its internal structure, rather than via a methodology that ensures the network is correct-by-construction. In this paper, we present a novel learning framework that takes an important first step towards realizing such a methodology. Our technique enables the construction of provably correct networks with respect to a broad class of safety properties, a capability that goes well-beyond existing approaches. Overcoming the challenge of general safety property enforcement within the network training process in a supervised learning pipeline, however, requires a fundamental shift in how we architect and build ANNs. Our key insight is that we can integrate an optimization-based abstraction refinement loop into the learning process that iteratively splits the input space from which training data is drawn, based on the efficacy with which such a partition enables safety verification. To do so, our approach enables training to take place over an abstraction of a concrete network that operates over dynamically constructed partitions of the input space. We provide theoretical results that show that classical gradient descent methods used to optimize these networks can be seamlessly adopted to this framework to ensure soundness of our approach. Moreover, we empirically demonstrate that realizing soundness does not come at the price of accuracy, giving us a meaningful pathway for building both precise and correct networks.

An Inductive Synthesis Framework for Verifiable Reinforcement Learning Artificial Intelligence

Despite the tremendous advances that have been made in the last decade on developing useful machine-learning applications, their wider adoption has been hindered by the lack of strong assurance guarantees that can be made about their behavior. In this paper, we consider how formal verification techniques developed for traditional software systems can be repurposed for verification of reinforcement learning-enabled ones, a particularly important class of machine learning systems. Rather than enforcing safety by examining and altering the structure of a complex neural network implementation, our technique uses blackbox methods to synthesizes deterministic programs, simpler, more interpretable, approximations of the network that can nonetheless guarantee desired safety properties are preserved, even when the network is deployed in unanticipated or previously unobserved environments. Our methodology frames the problem of neural network verification in terms of a counterexample and syntax-guided inductive synthesis procedure over these programs. The synthesis procedure searches for both a deterministic program and an inductive invariant over an infinite state transition system that represents a specification of an application's control logic. Additional specifications defining environment-based constraints can also be provided to further refine the search space. Synthesized programs deployed in conjunction with a neural network implementation dynamically enforce safety conditions by monitoring and preventing potentially unsafe actions proposed by neural policies. Experimental results over a wide range of cyber-physical applications demonstrate that software-inspired formal verification techniques can be used to realize trustworthy reinforcement learning systems with low overhead.