Attacks on Node Attributes in Graph Neural Networks

arXiv:2402.12426v1 Announce Type: cross
Abstract: Graphs are commonly used to model complex networks prevalent in modern social media and literacy applications. Our research investigates the vulnerability of these graphs through the application of feature based adversarial attacks, focusing on both decision-time attacks and poisoning attacks. In contrast to state-of-the-art models like Net Attack and Meta Attack, which target node attributes and graph structure, our study specifically targets node attributes. For our analysis, we utilized the text dataset Hellaswag and graph …

abstract adversarial adversarial attacks application applications art arxiv attacks contrast cs.ai cs.lg cs.si decision feature graph graph neural networks graphs literacy media modern networks neural networks node poisoning attacks research social social media state state-of-the-art models through type vulnerability