GraphLIME: Local Interpretable Model Explanations for Graph Neural Networks

Abstract

Graph structured data has wide applicability in various domains such as physics, chemistry, biology, computer vision, and social networks, to name a few. Recently, graph neural networks (GNN) were shown to be successful in effectively representing graph structured data because of their good performance and generalization ability. GNN is a deep learning based method that learns a node representation by combining specific nodes and the structural/topological information of a graph. However, like other deep models, explaining the effectiveness of GNN models is a challenging task because of the complex nonlinear transformations made over the iterations. In this paper, we propose GraphLIME, a local interpretable model explanation for graphs using the Hilbert-Schmidt Independence Criterion (HSIC) Lasso, which is a nonlinear feature selection method. GraphLIME is a generic GNN-model explanation framework that learns a nonlinear interpretable model locally in the subgraph of the node being explained. More specifically, to explain a node, we generate a nonlinear interpretable model from its $N$-hop neighborhood and then compute the K most representative features as the explanations of its prediction using HSIC Lasso. Through experiments on two real-world datasets, the explanations of GraphLIME are found to be of extraordinary degree and more descriptive in comparison to the existing explanation methods.

Cited by in corpus (13)

• Higher-Order Explanations of Graph Neural Networks via Relevant Walks
• A Comprehensive Survey on Trustworthy Graph Neural Networks: Privacy, Robustness, Fairness, and Explainability
• Alleviating Structural Distribution Shift in Graph Anomaly Detection
• SE-GSL: A General and Effective Graph Structure Learning Framework through Structural Entropy Optimization
• xFraud: Explainable Fraud Transaction Detection
• A Meta-Learning Approach for Training Explainable Graph Neural Networks
• Understanding the Message Passing in Graph Neural Networks via Power Iteration Clustering
• Visualizing Graph Neural Networks with CorGIE: Corresponding a Graph to Its Embedding
• View-based Explanations for Graph Neural Networks
• GRETEL: A unified framework for Graph Counterfactual Explanation Evaluation
• Private Graph Extraction via Feature Explanations
• Generating Post-hoc Explanations for Skip-gram-based Node Embeddings by Identifying Important Nodes with Bridgeness
• XInsight: Revealing Model Insights for GNNs with Flow-based Explanations