Refining Latent Homophilic Structures over Heterophilic Graphs for Robust Graph Convolution Networks

TL;DR

This paper introduces a novel method called LHS that enhances the robustness of Graph Convolutional Networks by learning and refining latent homophilic structures over heterophilic graphs, addressing structural out-of-distribution vulnerabilities.

Contribution

The paper pioneers a technique to automatically learn and refine latent homophilic structures in GCNs to improve robustness over heterophilic graphs.

Findings

LHS improves GCN robustness on heterophilic graphs.

The method effectively mitigates structural out-of-distribution issues.

Experimental results outperform existing approaches.

Abstract

Graph convolution networks (GCNs) are extensively utilized in various graph tasks to mine knowledge from spatial data. Our study marks the pioneering attempt to quantitatively investigate the GCN robustness over omnipresent heterophilic graphs for node classification. We uncover that the predominant vulnerability is caused by the structural out-of-distribution (OOD) issue. This finding motivates us to present a novel method that aims to harden GCNs by automatically learning Latent Homophilic Structures over heterophilic graphs. We term such a methodology as LHS. To elaborate, our initial step involves learning a latent structure by employing a novel self-expressive technique based on multi-node interactions. Subsequently, the structure is refined using a pairwisely constrained dual-view contrastive learning approach. We iteratively perform the above procedure, enabling a GCN model to aggregate information in a homophilic way on heterophilic graphs. Armed with such an adaptable structure, we can properly mitigate the structural OOD threats over heterophilic graphs. Experiments on various benchmarks show the effectiveness of the proposed LHS approach for robust GCNs.