Graph Contrastive Learning via Spectral Graph Alignment

TL;DR

This paper introduces SpecMatch-CL, a spectral graph alignment loss for contrastive learning that improves global structure control of graph embeddings, leading to state-of-the-art results on multiple benchmarks.

Contribution

The paper proposes a novel spectral alignment loss for contrastive graph learning, enhancing global structure control and improving performance over existing methods.

Findings

Achieves state-of-the-art results on eight TU benchmarks.

Yields consistent gains in transfer learning on PPI-306K and ZINC 2M datasets.

Provides theoretical bounds relating spectral Laplacian differences to contrastive loss quality.

Abstract

Given augmented views of each input graph, contrastive learning methods (e.g., InfoNCE) optimize pairwise alignment of graph embeddings across views while providing no mechanism to control the global structure of the view specific graph-of-graphs built from these embeddings. We introduce SpecMatch-CL, a novel loss function that aligns the view specific graph-of-graphs by minimizing the difference between their normalized Laplacians. Theoretically, we show that under certain assumptions, the difference between normalized Laplacians provides an upper bound not only for the difference between the ideal Perfect Alignment contrastive loss and the current loss, but also for the Uniformly loss. Empirically, SpecMatch-CL establishes new state of the art on eight TU benchmarks under unsupervised learning and semi-supervised learning at low label rates, and yields consistent gains in transfer learning on PPI-306K and ZINC 2M datasets.