Unifying Graph Contrastive Learning via Graph Message Augmentation

TL;DR

The work introduces Graph Message Augmentation (GMA) as a universal framework to unify graph data augmentations by operating on graph messages, enabling straightforward graph-mixup and broader applicability across graph types. Building on GMA, it proposes Graph Message Contrastive Learning (GMCL) with an adaptive, attribution-guided augmenter (AttGMA) to preserve label-invariant information during augmentation. Empirical results across unsupervised, semi-supervised, transfer, and robustness tests show GMCL, particularly with AttGMA, achieves superior performance, generalizability, and stability compared to prior GCL methods. The approach offers a principled, scalable pathway to unify augmentation design in graph representation learning and improve practical performance.

Abstract

Graph contrastive learning is usually performed by first conducting Graph Data Augmentation (GDA) and then employing a contrastive learning pipeline to train GNNs. As we know that GDA is an important issue for graph contrastive learning. Various GDAs have been developed recently which mainly involve dropping or perturbing edges, nodes, node attributes and edge attributes. However, to our knowledge, it still lacks a universal and effective augmentor that is suitable for different types of graph data. To address this issue, in this paper, we first introduce the graph message representation of graph data. Based on it, we then propose a novel Graph Message Augmentation (GMA), a universal scheme for reformulating many existing GDAs. The proposed unified GMA not only gives a new perspective to understand many existing GDAs but also provides a universal and more effective graph data augmentation for graph self-supervised learning tasks. Moreover, GMA introduces an easy way to implement the mixup augmentor which is natural for images but usually challengeable for graphs. Based on the proposed GMA, we then propose a unified graph contrastive learning, termed Graph Message Contrastive Learning (GMCL), that employs attribution-guided universal GMA for graph contrastive learning. Experiments on many graph learning tasks demonstrate the effectiveness and benefits of the proposed GMA and GMCL approaches.

Figures (8)

• Figure 1: Some comparison performance of different GDAs and our GMA on bioinformatics (MUTAG), citation (Citeseer) and molecular (MUV) datasets. GMA performs consistently better than traditional GDAs.
• Figure 2: Overview of three kinds of regular GCL frameworks and the proposed GMCL framework.
• Figure 3: (a) An example of the graph with its message representation Eq.(9). (b-f) Demonstration of different GDAs and corresponding GMA forms.
• Figure 4: The pipeline of GMA via graph message mixup.
• Figure 5: The architecture of Graph Message Contrastive Learning (GMCL) framework for graph classification tasks.