Negative Metric Learning for Graphs

TL;DR

This paper introduces NML-GCL, a novel graph contrastive learning approach that uses a learnable negative metric to better distinguish false negatives, improving downstream task performance.

Contribution

It proposes a Negative Metric Network with bi-level optimization for self-supervised false negative detection in graph contrastive learning.

Findings

Outperforms existing GCL methods on benchmark datasets

Theoretical analysis supports the effectiveness of NML-GCL

Extensive experiments demonstrate improved downstream task results

Abstract

Graph contrastive learning (GCL) often suffers from false negatives, which degrades the performance on downstream tasks. The existing methods addressing the false negative issue usually rely on human prior knowledge, still leading GCL to suboptimal results. In this paper, we propose a novel Negative Metric Learning (NML) enhanced GCL (NML-GCL). NML-GCL employs a learnable Negative Metric Network (NMN) to build a negative metric space, in which false negatives can be distinguished better from true negatives based on their distance to anchor node. To overcome the lack of explicit supervision signals for NML, we propose a joint training scheme with bi-level optimization objective, which implicitly utilizes the self-supervision signals to iteratively optimize the encoder and the negative metric network. The solid theoretical analysis and the extensive experiments conducted on widely used benchmarks verify the superiority of the proposed method.