GNN-Explainer

GNN-Explainer is the first general tool for explaining predictions made by graph neural networks (GNNs). Given a trained GNN model and an instance as its input, the GNN-Explainer produces an explanation of the GNN model prediction via a compact subgraph structure, as well as a set of feature dimensions important for its prediction. Graph Neural Networks (GNNs) are a powerful tool for machine learning on graphs. GNNs combine node feature information with the graph structure by recursively passing neural messages along edges of the input graph. However, incorporating both graph structure and feature information leads to complex models and explaining predictions made by GNNs remains unsolved.