Adversarial Graph Augmentation to Improve Graph Contrastive Learning
This work addresses label scarcity in real-world graph data by improving graph contrastive learning, offering incremental gains in performance for tasks like molecule property prediction and social network classification.
The paper tackles the problem of graph neural networks capturing redundant features in graph contrastive learning, which leads to subpar downstream performance, by proposing adversarial graph augmentation to avoid redundancy and achieving performance gains of up to 14% in unsupervised, 6% in transfer, and 3% in semi-supervised learning across 18 benchmark datasets.
Self-supervised learning of graph neural networks (GNN) is in great need because of the widespread label scarcity issue in real-world graph/network data. Graph contrastive learning (GCL), by training GNNs to maximize the correspondence between the representations of the same graph in its different augmented forms, may yield robust and transferable GNNs even without using labels. However, GNNs trained by traditional GCL often risk capturing redundant graph features and thus may be brittle and provide sub-par performance in downstream tasks. Here, we propose a novel principle, termed adversarial-GCL (AD-GCL), which enables GNNs to avoid capturing redundant information during the training by optimizing adversarial graph augmentation strategies used in GCL. We pair AD-GCL with theoretical explanations and design a practical instantiation based on trainable edge-dropping graph augmentation. We experimentally validate AD-GCL by comparing with the state-of-the-art GCL methods and achieve performance gains of up-to $14\%$ in unsupervised, $6\%$ in transfer, and $3\%$ in semi-supervised learning settings overall with 18 different benchmark datasets for the tasks of molecule property regression and classification, and social network classification.