Discrepancy-Aware Graph Mask Auto-Encoder

TL;DR

This paper introduces DGMAE, a novel graph auto-encoder that captures discrepancy information between nodes, improving representation learning especially for heterophilic graphs, and outperforms existing methods on multiple benchmarks.

Contribution

It proposes a discrepancy-aware auto-encoder that reconstructs node discrepancies, enhancing distinguishability in heterophilic graphs, a novel approach in graph self-supervised learning.

Findings

DGMAE outperforms state-of-the-art methods on node classification, clustering, and graph classification.

It effectively preserves node discrepancies in low-dimensional embeddings.

Experimental results on 17 benchmark datasets validate its superiority.

Abstract

Masked Graph Auto-Encoder, a powerful graph self-supervised training paradigm, has recently shown superior performance in graph representation learning. Existing works typically rely on node contextual information to recover the masked information. However, they fail to generalize well to heterophilic graphs where connected nodes may be not similar, because they focus only on capturing the neighborhood information and ignoring the discrepancy information between different nodes, resulting in indistinguishable node representations. In this paper, to address this issue, we propose a Discrepancy-Aware Graph Mask Auto-Encoder (DGMAE). It obtains more distinguishable node representations by reconstructing the discrepancy information of neighboring nodes during the masking process. We conduct extensive experiments on 17 widely-used benchmark datasets. The results show that our DGMAE can effectively preserve the discrepancies of nodes in low-dimensional space. Moreover, DGMAE significantly outperforms state-of-the-art graph self-supervised learning methods on three graph analytic including tasks node classification, node clustering, and graph classification, demonstrating its remarkable superiority. The code of DGMAE is available at https://github.com/zhengziyu77/DGMAE.