Bootstrapping Informative Graph Augmentation via A Meta Learning Approach
This addresses the issue of non-learnable augmentations in graph self-supervised learning, offering a novel method for generating more informative augmentations, though it appears incremental as it builds on existing contrastive learning frameworks.
The paper tackles the problem of generating unbeneficial augmented graphs in graph contrastive learning by proposing a learnable graph augmenter called MEGA, which improves representation learning and outperforms state-of-the-art methods on multiple benchmark datasets.
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.