Explaining Graph Neural Networks via Structure-aware Interaction Index
This work addresses the need for better interpretability tools for graph neural networks, which is crucial for users in fields like bioinformatics or social network analysis, though it appears incremental as it builds on existing Shapley-based approaches.
The paper tackled the problem of explaining graph neural networks by introducing the Myerson-Taylor interaction index, which incorporates graph structure into attributing node and interaction values, and demonstrated that their method, MAGE, consistently provides superior subgraph explanations compared to state-of-the-art methods in experiments on various datasets and models.
The Shapley value is a prominent tool for interpreting black-box machine learning models thanks to its strong theoretical foundation. However, for models with structured inputs, such as graph neural networks, existing Shapley-based explainability approaches either focus solely on node-wise importance or neglect the graph structure when perturbing the input instance. This paper introduces the Myerson-Taylor interaction index that internalizes the graph structure into attributing the node values and the interaction values among nodes. Unlike the Shapley-based methods, the Myerson-Taylor index decomposes coalitions into components satisfying a pre-chosen connectivity criterion. We prove that the Myerson-Taylor index is the unique one that satisfies a system of five natural axioms accounting for graph structure and high-order interaction among nodes. Leveraging these properties, we propose Myerson-Taylor Structure-Aware Graph Explainer (MAGE), a novel explainer that uses the second-order Myerson-Taylor index to identify the most important motifs influencing the model prediction, both positively and negatively. Extensive experiments on various graph datasets and models demonstrate that our method consistently provides superior subgraph explanations compared to state-of-the-art methods.