Deep Graph Contrastive Representation Learning

TLDR

Graph representation learning is essential for analyzing graph‑structured data. The paper proposes an unsupervised graph representation learning framework that employs a node‑level contrastive objective. The framework generates two corrupted graph views, learns node representations by maximizing agreement between them, uses a hybrid scheme to perturb structure and attributes, and is theoretically justified via mutual information and triplet loss. Experiments show the method consistently outperforms state‑of‑the‑art baselines and even surpasses supervised counterparts on transductive tasks.

Abstract

Graph representation learning nowadays becomes fundamental in analyzing graph-structured data. Inspired by recent success of contrastive methods, in this paper, we propose a novel framework for unsupervised graph representation learning by leveraging a contrastive objective at the node level. Specifically, we generate two graph views by corruption and learn node representations by maximizing the agreement of node representations in these two views. To provide diverse node contexts for the contrastive objective, we propose a hybrid scheme for generating graph views on both structure and attribute levels. Besides, we provide theoretical justification behind our motivation from two perspectives, mutual information and the classical triplet loss. We perform empirical experiments on both transductive and inductive learning tasks using a variety of real-world datasets. Experimental experiments demonstrate that despite its simplicity, our proposed method consistently outperforms existing state-of-the-art methods by large margins. Moreover, our unsupervised method even surpasses its supervised counterparts on transductive tasks, demonstrating its great potential in real-world applications.