DAG Matters! GFlowNets Enhanced Explainer For Graph Neural Networks
This addresses the challenge of providing faithful explanations for GNN predictions in large-scale settings, which is an incremental improvement over existing combinatorial optimization methods.
The paper tackles the problem of explaining predictions of graph neural networks (GNNs) by proposing GFlowExplainer, a generative method that turns subgraph selection into a step-by-step process, eliminating the need for pre-training and enabling scalability to large graphs, with experiments showing superior performance on synthetic and real datasets.
Uncovering rationales behind predictions of graph neural networks (GNNs) has received increasing attention over the years. Existing literature mainly focus on selecting a subgraph, through combinatorial optimization, to provide faithful explanations. However, the exponential size of candidate subgraphs limits the applicability of state-of-the-art methods to large-scale GNNs. We enhance on this through a different approach: by proposing a generative structure -- GFlowNets-based GNN Explainer (GFlowExplainer), we turn the optimization problem into a step-by-step generative problem. Our GFlowExplainer aims to learn a policy that generates a distribution of subgraphs for which the probability of a subgraph is proportional to its' reward. The proposed approach eliminates the influence of node sequence and thus does not need any pre-training strategies. We also propose a new cut vertex matrix to efficiently explore parent states for GFlowNets structure, thus making our approach applicable in a large-scale setting. We conduct extensive experiments on both synthetic and real datasets, and both qualitative and quantitative results show the superiority of our GFlowExplainer.