Instance Discrimination for Link Prediction

TL;DR

This paper adapts instance discrimination models for link prediction in graphs, introduces structural augmentation based on community structure, and proposes two new models, L-GRACE and L-BGRL, achieving competitive results.

Contribution

It introduces link-based instance discrimination models and a community-structure augmentation for improved link prediction performance.

Findings

L-GRACE and L-BGRL outperform existing methods on unattributed graphs.

Performance depends heavily on augmentation strategies.

Proposed models match state-of-the-art results in supervised and self-supervised settings.

Abstract

Recently, instance discrimination models have emerged as a major solution for self-supervised learning. Having already demonstrated its effectiveness in the image domain, instance discrimination learning is now proving equally convincing in the graph domain, in particular for node classification. However, fewer contributions have tackled the link prediction task. In this contribution, we propose to adapt existing methods to this context. We first provide a rigorous evaluation of existing self-supervised models in the field of link prediction, showing that the main performance depends on the augmentation process (like in computer vision). We then propose a new structural augmentation based on the community structure that is relevant for link prediction. Our main contribution introduces two new models, L-GRACE and L-BGRL, based on link representations instead of node representations, which improve the performance of the existing methods, especially on unattributed graphs, and we show that they perform on par with the state of the art, both in supervised and self-supervised contexts.