Bootstrapping Informative Graph Augmentation via A Meta Learning Approach

TL;DR

This paper introduces MEGA, a learnable graph augmentation method using meta-learning to generate informative and uniform augmentations, improving graph contrastive learning performance.

Contribution

The paper proposes a novel meta-learning based graph augmenter that ensures uniformity and informativeness, enhancing self-supervised graph representation learning.

Findings

MEGA outperforms state-of-the-art methods on multiple benchmarks.

The learned augmentation improves the discriminative power of graph representations.

Different components of MEGA contribute significantly to its effectiveness.

Abstract

Recent works explore learning graph representations in a self-supervised manner. In graph contrastive learning, benchmark methods apply various graph augmentation approaches. However, most of the augmentation methods are non-learnable, which causes the issue of generating unbeneficial augmented graphs. Such augmentation may degenerate the representation ability of graph contrastive learning methods. Therefore, we motivate our method to generate augmented graph by a learnable graph augmenter, called MEta Graph Augmentation (MEGA). We then clarify that a "good" graph augmentation must have uniformity at the instance-level and informativeness at the feature-level. To this end, we propose a novel approach to learning a graph augmenter that can generate an augmentation with uniformity and informativeness. The objective of the graph augmenter is to promote our feature extraction network to learn a more discriminative feature representation, which motivates us to propose a meta-learning paradigm. Empirically, the experiments across multiple benchmark datasets demonstrate that MEGA outperforms the state-of-the-art methods in graph self-supervised learning tasks. Further experimental studies prove the effectiveness of different terms of MEGA.