English

Towards Generalizable Graph Contrastive Learning: An Information Theory Perspective

Machine Learning 2022-11-22 v1 Artificial Intelligence

Abstract

Graph contrastive learning (GCL) emerges as the most representative approach for graph representation learning, which leverages the principle of maximizing mutual information (InfoMax) to learn node representations applied in downstream tasks. To explore better generalization from GCL to downstream tasks, previous methods heuristically define data augmentation or pretext tasks. However, the generalization ability of GCL and its theoretical principle are still less reported. In this paper, we first propose a metric named GCL-GE for GCL generalization ability. Considering the intractability of the metric due to the agnostic downstream task, we theoretically prove a mutual information upper bound for it from an information-theoretic perspective. Guided by the bound, we design a GCL framework named InfoAdv with enhanced generalization ability, which jointly optimizes the generalization metric and InfoMax to strike the right balance between pretext task fitting and the generalization ability on downstream tasks. We empirically validate our theoretical findings on a number of representative benchmarks, and experimental results demonstrate that our model achieves state-of-the-art performance.

Keywords

Cite

@article{arxiv.2211.10929,
  title  = {Towards Generalizable Graph Contrastive Learning: An Information Theory Perspective},
  author = {Yige Yuan and Bingbing Xu and Huawei Shen and Qi Cao and Keting Cen and Wen Zheng and Xueqi Cheng},
  journal= {arXiv preprint arXiv:2211.10929},
  year   = {2022}
}

Comments

25 pages, 7 figures, 6 tables

R2 v1 2026-06-28T06:18:14.379Z