Turning Strengths Into Weaknesses: A Certified Robustness Inspired Attack Framework Against Graph Neural Networks

Binghui Wang, Meng Pang, Yun Dong; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023, pp. 16394-16403

Graph neural networks (GNNs) have achieved state-of-the-art performance in many graph-related tasks such as node classification. However, recent studies show that GNNs are vulnerable to both test-time and training-time attacks that perturb the graph structure. While the existing attack methods have shown promising attack performance, we would like to design an attack framework that can significantly enhance both the existing evasion and poisoning attacks. In particular, our attack framework is inspired by certified robustness. Certified robustness was originally used by defenders to defend against adversarial attacks. We are the first, from the attacker perspective, to leverage its properties to better attack GNNs. Specifically, we first leverage and derive nodes' certified perturbation sizes against evasion and poisoning attacks based on randomized smoothing. A larger certified perturbation size of a node indicates this node is theoretically more robust to graph perturbations. Such a property motivates us to focus more on nodes with smaller certified perturbation sizes, as they are easier to be attacked after graph perturbations. Accordingly, we design a certified robustness inspired attack loss, when incorporated into (any) existing attacks, produces our certified robustness inspired attack framework. We apply our attack framework to the existing attacks and results show it can significantly enhance the existing attacks' performance.