Towards Robust Graph Contrastive Learning

Abstract

We study the problem of adversarially robust self-supervised learning on graphs. In the contrastive learning framework, we introduce a new method that increases the adversarial robustness of the learned representations through i) adversarial transformations and ii) transformations that not only remove but also insert edges. We evaluate the learned representations in a preliminary set of experiments, obtaining promising results. We believe this work takes an important step towards incorporating robustness as a viable auxiliary task in graph contrastive learning.

References in corpus (6)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Fast Graph Representation Learning with PyTorch Geometric
• Deep Graph Contrastive Representation Learning
• DeepRobust: A PyTorch Library for Adversarial Attacks and Defenses
• Pre-Training Graph Neural Networks for Generic Structural Feature Extraction
• Self-supervised Graph Representation Learning via Bootstrapping