Reliable Graph Neural Network Explanations Through Adversarial Training

Abstract

Graph neural network (GNN) explanations have largely been facilitated through post-hoc introspection. While this has been deemed successful, many post-hoc explanation methods have been shown to fail in capturing a model's learned representation. Due to this problem, it is worthwhile to consider how one might train a model so that it is more amenable to post-hoc analysis. Given the success of adversarial training in the computer vision domain to train models with more reliable representations, we propose a similar training paradigm for GNNs and analyze the respective impact on a model's explanations. In instances without ground truth labels, we also determine how well an explanation method is utilizing a model's learned representation through a new metric and demonstrate adversarial training can help better extract domain-relevant insights in chemistry.

arXiv comment4 pages, 3 figures, ICML Workshop on Theoretic Foundation, Criticism, and Application Trend of Explainable AI

References in corpus (10)

• Semi-Supervised Classification with Graph Convolutional Networks
• Neural Message Passing for Quantum Chemistry
• A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAI
• Graph Neural Network for Traffic Forecasting: A Survey
• Adversarial Examples Are Not Bugs, They Are Features
• Convolutional Neural Network Architectures for Signals Supported on Graphs
• Higher-Order Explanations of Graph Neural Networks via Relevant Walks
• PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networks
• Generative Counterfactual Introspection for Explainable Deep Learning
• Generative Causal Explanations for Graph Neural Networks