Membership Inference Attacks Against Robust Graph Neural Network

Liu, Zhengyang, Zhang, Xiaoyu, Chen, Chenyang, Lin, Shen, and Li, Jingjin (2022) Membership Inference Attacks Against Robust Graph Neural Network. In: Cyberspace Safety and Security, CSS 2022 (13547) pp. 259-273. From: Cyberspace Safety and Security: 14th International Symposium, CSS 2022, 16–18 October 2022, Xi’an, China.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only

View at Publisher Website: https://doi.org/10.1007/978-3-031-18067-...
 
1
1


Abstract

With the rapid development of neural network technologies in machine learning, neural networks are widely used in artificial intelligence tasks. Due to the widespread existence of graph data, graph neural networks, a kind of neural network specializing in processing graph data, has become a research hotspot. This paper firstly studies the relationship between adversarial attacks and privacy attacks on graphs, i.e., whether a robust model trained on graph adversarial can improve the attack effect of graph membership inference attacks. We also find the different performance of the robust model’s loss function on the training set and the test set is a critical reason for the increasing membership inference attack success rate. Extensive experimental evaluations on Cora, Cora-ml, Citeseer, Polblogs and Pubmed demonstrate that the robust model obtained by adversarial training can significantly improve the attack success rate of membership inference attacks.

Item ID: 77164
Item Type: Conference Item (Research - E1)
ISBN: 978-3-031-18067-5
ISSN: 1611-3349
Keywords: Graph neural network, Membership inference attack, Robust model, Adversarial training
Copyright Information: © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2022.
Date Deposited: 04 Jan 2023 09:01
FoR Codes: 46 INFORMATION AND COMPUTING SCIENCES > 4604 Cybersecurity and privacy > 460407 System and network security @ 100%
SEO Codes: 22 INFORMATION AND COMMUNICATION SERVICES > 2204 Information systems, technologies and services > 220405 Cybersecurity @ 100%
Downloads: Total: 1
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page