http://www.ece.virginia.edu/~jl6qk/pubs/ECMLPKDD2024.pdf WebNov 19, 2024 · Under these scenarios, exploiting GNN's vulnerabilities and further downgrading its performance become extremely incentive for adversaries. Previous attackers mainly focus on structural perturbations or node injections to the existing graphs, guided by gradients from the surrogate models.
Hard Label Black Box Node Injection Attack on Graph Neural …
WebIn this paper, we develop a more scalable framework named Approximate Fast Gradient Sign Method which considers a more practical attack scenario where adversaries can … WebApr 5, 2024 · Rethinking the Trigger-injecting Position in Graph Backdoor Attack. Jing Xu, Gorka Abad, Stjepan Picek. Published 5 April 2024. Computer Science. Backdoor attacks have been demonstrated as a security threat for machine learning models. Traditional backdoor attacks intend to inject backdoor functionality into the model such that the … orangeolive repair
Understanding and Improving Graph Injection Attack by …
WebSep 1, 2024 · A more realistic scenario, graph injection attack (GIA), is studied in [14, 13], which injects new vicious nodes instead of modifying the original graph. A greedy … WebApr 7, 2024 · State estimation plays a vital role in the stable operation of modern power systems, but it is vulnerable to cyber attacks. False data injection attacks (FDIA), one of the most common cyber attacks, can tamper with measurement data and bypass the bad data detection (BDD) mechanism, leading to incorrect results of power system state … WebApr 20, 2024 · Hence, we consider a novel form of node injection poisoning attacks on graph data. We model the key steps of a node injection attack, e.g., establishing links between the injected adversarial nodes and other nodes, choosing the label of an injected node, etc. by a Markov Decision Process. iphonexmax多大