Timezone: »

GNNExplainer: Generating Explanations for Graph Neural Networks
Zhitao Ying · Dylan Bourgeois · Jiaxuan You · Marinka Zitnik · Jure Leskovec

Tue Dec 10 10:45 AM -- 12:45 PM (PST) @ East Exhibition Hall B + C #134

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. Here we propose GNNExplainer, the first general, model-agnostic approach for providing interpretable explanations for predictions of any GNN-based model on any graph-based machine learning task. Given an instance, GNNExplainer identifies a compact subgraph structure and a small subset of node features that have a crucial role in GNN's prediction. Further, GNNExplainer can generate consistent and concise explanations for an entire class of instances. We formulate GNNExplainer as an optimization task that maximizes the mutual information between a GNN's prediction and distribution of possible subgraph structures. Experiments on synthetic and real-world graphs show that our approach can identify important graph structures as well as node features, and outperforms baselines by 17.1% on average. GNNExplainer provides a variety of benefits, from the ability to visualize semantically relevant structures to interpretability, to giving insights into errors of faulty GNNs.

Author Information

Zhitao Ying (Stanford University)
Dylan Bourgeois (EPFL)
Jiaxuan You (Stanford University)

2nd CS PhD student in Stanford

Marinka Zitnik (Harvard University)
Jure Leskovec (Stanford University and Pinterest)

More from the Same Authors