SEGA: Structural Entropy Guided Anchor View for Graph Contrastive Learning

TL;DR

SEGA introduces a novel anchor view based on structural entropy for graph contrastive learning, preserving essential information and improving performance across multiple benchmarks.

Contribution

The paper proposes a new anchor view guided by structural entropy, grounded in graph information bottleneck theory, to enhance graph contrastive learning.

Findings

Significant performance improvements on graph classification benchmarks.

Effective preservation of essential graph information.

Versatility across unsupervised, semi-supervised, and transfer learning settings.

Abstract

In contrastive learning, the choice of ``view'' controls the information that the representation captures and influences the performance of the model. However, leading graph contrastive learning methods generally produce views via random corruption or learning, which could lead to the loss of essential information and alteration of semantic information. An anchor view that maintains the essential information of input graphs for contrastive learning has been hardly investigated. In this paper, based on the theory of graph information bottleneck, we deduce the definition of this anchor view; put differently, \textit{the anchor view with essential information of input graph is supposed to have the minimal structural uncertainty}. Furthermore, guided by structural entropy, we implement the anchor view, termed \textbf{SEGA}, for graph contrastive learning. We extensively validate the proposed anchor view on various benchmarks regarding graph classification under unsupervised, semi-supervised, and transfer learning and achieve significant performance boosts compared to the state-of-the-art methods.