DEGREE: Decomposition Based Explanation For Graph Neural Networks
This addresses the interpretability challenge for GNNs, which hampers trust and applicability in graph data applications, offering a novel approach to improve explanation faithfulness.
The paper tackles the problem of explaining Graph Neural Networks (GNNs) by proposing DEGREE, a decomposition-based method that provides faithful explanations for predictions, demonstrating effectiveness on node and graph classification tasks with quantitative and qualitative experiments.
Graph Neural Networks (GNNs) are gaining extensive attention for their application in graph data. However, the black-box nature of GNNs prevents users from understanding and trusting the models, thus hampering their applicability. Whereas explaining GNNs remains a challenge, most existing methods fall into approximation based and perturbation based approaches with suffer from faithfulness problems and unnatural artifacts, respectively. To tackle these problems, we propose DEGREE \degree to provide a faithful explanation for GNN predictions. By decomposing the information generation and aggregation mechanism of GNNs, DEGREE allows tracking the contributions of specific components of the input graph to the final prediction. Based on this, we further design a subgraph level interpretation algorithm to reveal complex interactions between graph nodes that are overlooked by previous methods. The efficiency of our algorithm can be further improved by utilizing GNN characteristics. Finally, we conduct quantitative and qualitative experiments on synthetic and real-world datasets to demonstrate the effectiveness of DEGREE on node classification and graph classification tasks.