[KDD 2020] Graph Structure Learning for Robust Graph Neural Networks
Aug 13, 202010 views
Graph Neural Networks (GNNs) are powerful tools in representation learning for graphs. However, recent studies show that GNNs,are vulnerable to carefully-crafted perturbations, called adversarial,attacks. Adversarial attacks can easily fool GNNs in making predictions for downstream tasks. The vulnerability to adversarial attacks,has raised increasing concerns for applying GNNs in safety-critical,applications. Therefore, developing robust algorithms to defend,adversarial attacks is of great signi,,cance. A natural idea to defend,adversarial attacks is to clean the perturbed graph. It is evident,that real-world graphs share some intrinsic properties. For example,,many real-world graphs are low-rank and sparse, and the features of,two adjacent nodes tend to be similar. In fact, we ,,nd that adversarial attacks are likely to violate these graph properties. Therefore, in,this paper, we explore these properties to defend adversarial attacks,on graphs. In particular, we propose a general framework Pro-GNN,,which can jointly learn a structural graph and a robust graph neural,network model from the perturbed graph guided by these properties. Extensive experiments on real-world graphs demonstrate that,the proposed framework achieves signi,,cantly better performance,compared with the state-of-the-art defense methods, even when,the graph is heavily perturbed. We release the implementation of,Pro-GNN to our DeepRobust repository for adversarial attacks and,defenses,1,. The speci,,c experimental settings to reproduce our results can be found in https://github.com/ChandlerBang/Pro-GNN.