PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networks

Abstract

In Graph Neural Networks (GNNs), the graph structure is incorporated into the learning of node representations. This complex structure makes explaining GNNs' predictions become much more challenging. In this paper, we propose PGM-Explainer, a Probabilistic Graphical Model (PGM) model-agnostic explainer for GNNs. Given a prediction to be explained, PGM-Explainer identifies crucial graph components and generates an explanation in form of a PGM approximating that prediction. Different from existing explainers for GNNs where the explanations are drawn from a set of linear functions of explained features, PGM-Explainer is able to demonstrate the dependencies of explained features in form of conditional probabilities. Our theoretical analysis shows that the PGM generated by PGM-Explainer includes the Markov-blanket of the target prediction, i.e. including all its statistical information. We also show that the explanation returned by PGM-Explainer contains the same set of independence statements in the perfect map. Our experiments on both synthetic and real-world datasets show that PGM-Explainer achieves better performance than existing explainers in many benchmark tasks.

References in corpus (10)

• Adam: A Method for Stochastic Optimization
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Inductive Representation Learning on Large Graphs
• Towards A Rigorous Science of Interpretable Machine Learning
• Learning Important Features Through Propagating Activation Differences
• Modeling polypharmacy side effects with graph convolutional networks
• Self-Attention Graph Pooling
• Link Prediction Based on Graph Neural Networks
• Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation
• Benchmarking Graph Neural Networks

Cited by in corpus (7)

• Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future
• A Comprehensive Survey on Trustworthy Graph Neural Networks: Privacy, Robustness, Fairness, and Explainability
• Explaining the Explainers in Graph Neural Networks: a Comparative Study
• GNNShap: Scalable and Accurate GNN Explanation using Shapley Values
• Generating Post-hoc Explanations for Skip-gram-based Node Embeddings by Identifying Important Nodes with Bridgeness
• Reliable Graph Neural Network Explanations Through Adversarial Training
• SEEN: Sharpening Explanations for Graph Neural Networks using Explanations from Neighborhoods