A Hard Label Black-box Adversarial Attack Against Graph Neural Networks
Jiaming Mu, Binghui Wang, Qi Li, Kun Sun, Mingwei Xu, Zhuotao Liu
Abstract
Graph Neural Networks (GNNs) have achieved state-of-the-art performance in various graph structure related tasks such as node classification and graph classification. However, GNNs are vulnerable to adversarial attacks. Existing works mainly focus on attacking GNNs for node classification; nevertheless, the attacks against GNNs for graph classification have not been well explored.
Topics & Concepts
Computer scienceAdversarial systemGraphNode (physics)Artificial neural networkArtificial intelligenceMachine learningTheoretical computer scienceEngineeringStructural engineeringAdversarial Robustness in Machine LearningAdvanced Graph Neural NetworksMachine Learning in Materials Science