Towards Graph Contrastive Learning: A Survey and Beyond

TL;DR

This paper surveys Graph Contrastive Learning (GCL), a key component of self-supervised learning on graphs, discussing its principles, extensions, applications, challenges, and future directions to advance data-efficient graph learning.

Contribution

It provides the first comprehensive survey dedicated to GCL, covering fundamental principles, extensions, applications, and future challenges in the field.

Findings

GCL is effective in various data-efficient graph learning scenarios.

Extensions of GCL improve performance in weakly supervised and transfer learning.

GCL has broad applications in drug discovery, genomics, and recommender systems.

Abstract

In recent years, deep learning on graphs has achieved remarkable success in various domains. However, the reliance on annotated graph data remains a significant bottleneck due to its prohibitive cost and time-intensive nature. To address this challenge, self-supervised learning (SSL) on graphs has gained increasing attention and has made significant progress. SSL enables machine learning models to produce informative representations from unlabeled graph data, reducing the reliance on expensive labeled data. While SSL on graphs has witnessed widespread adoption, one critical component, Graph Contrastive Learning (GCL), has not been thoroughly investigated in the existing literature. Thus, this survey aims to fill this gap by offering a dedicated survey on GCL. We provide a comprehensive overview of the fundamental principles of GCL, including data augmentation strategies, contrastive modes, and contrastive optimization objectives. Furthermore, we explore the extensions of GCL to other aspects of data-efficient graph learning, such as weakly supervised learning, transfer learning, and related scenarios. We also discuss practical applications spanning domains such as drug discovery, genomics analysis, recommender systems, and finally outline the challenges and potential future directions in this field.