ES-GNN: Generalizing Graph Neural Networks Beyond Homophily with Edge Splitting

Abstract

While Graph Neural Networks (GNNs) have achieved enormous success in multiple graph analytical tasks, modern variants mostly rely on the strong inductive bias of homophily. However, real-world networks typically exhibit both homophilic and heterophilic linking patterns, wherein adjacent nodes may share dissimilar attributes and distinct labels. Therefore, GNNs smoothing node proximity holistically may aggregate both task-relevant and irrelevant (even harmful) information, limiting their ability to generalize to heterophilic graphs and potentially causing non-robustness. In this work, we propose a novel Edge Splitting GNN (ES-GNN) framework to adaptively distinguish between graph edges either relevant or irrelevant to learning tasks. This essentially transfers the original graph into two subgraphs with the same node set but complementary edge sets dynamically. Given that, information propagation separately on these subgraphs and edge splitting are alternatively conducted, thus disentangling the task-relevant and irrelevant features. Theoretically, we show that our ES-GNN can be regarded as a solution to a disentangled graph denoising problem, which further illustrates our motivations and interprets the improved generalization beyond homophily. Extensive experiments over 11 benchmark and 1 synthetic datasets not only demonstrate the effective performance of ES-GNN but also highlight its robustness to adversarial graphs and mitigation of the over-smoothing problem.

References in corpus (14)

• Graph Neural Networks with convolutional ARMA filters
• Improving Graph Neural Network Expressivity via Subgraph Isomorphism Counting
• Higher-Order Explanations of Graph Neural Networks via Relevant Walks
• Disentangled Contrastive Collaborative Filtering
• Multi-view Intent Disentangle Graph Networks for Bundle Recommendation
• Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns
• Reinforced Causal Explainer for Graph Neural Networks
• Context Disentangling and Prototype Inheriting for Robust Visual Grounding
• Image-to-Image Translation with Disentangled Latent Vectors for Face Editing
• Exploring Edge Disentanglement for Node Classification
• Interpretable Deep Graph Generation with Node-Edge Co-Disentanglement
• Meta-Weight Graph Neural Network: Push the Limits Beyond Global Homophily
• Soft-mask: Adaptive Substructure Extractions for Graph Neural Networks
• Synthetic Graph Generation to Benchmark Graph Learning