跳到主要导航 跳到搜索 跳到主要内容

GNNHacker: Adaptive Subgraph Backdoor Attacking Method with Saliency Analysis and Joint Optimization

  • Xiaojie Wu
  • , Yan Luo
  • , Shujie Li
  • , Xiping Hu
  • , Qiang Liu*
  • , Chengming Li*
  • *此作品的通讯作者
  • National University of Defense Technology
  • Hong Kong University of Science and Technology
  • University of Science and Technology of China
  • Shenzhen MSU-BIT University

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

Subgraph backdoor attacks reveal critical vulnerabilities in graph neural networks (GNNs). They replace GNNs’ nodes and edges with elaborate triggers, causing the misclassification of graph-structured data examples into target categories specified by adversaries. To overcome the high computational overhead of existing attacking methods, we propose an adaptive subgraph backdoor attack method with saliency analysis and joint optimization called GNNHacker. Specifically, GNNHacker introduces a gradient-based saliency map to identify the nodes with high saliency scores as poisoning targets. Then, it adopts a joint optimization mechanism by simultaneously optimizing trigger generation and the training of backdoor GNNs. Experimental results over five public datasets show that the proposed method significantly outperforms baselines in terms of attacking capability and efficiency.

源语言英语
主期刊名Advanced Intelligent Computing Technology and Applications - 21st International Conference, ICIC 2025, Proceedings
编辑De-Shuang Huang, Yijie Pan, Wei Chen, Haiming Chen
出版商Springer Science and Business Media Deutschland GmbH
27-43
页数17
ISBN(印刷版)9789819698486
DOI
出版状态已出版 - 2025
已对外发布
活动21st International Conference on Intelligent Computing, ICIC 2025 - Ningbo, 中国
期限: 26 7月 202529 7月 2025

丛书

姓名Lecture Notes in Computer Science
15851 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议21st International Conference on Intelligent Computing, ICIC 2025
国家/地区中国
Ningbo
时期26/07/2529/07/25

学术指纹

探究 'GNNHacker: Adaptive Subgraph Backdoor Attacking Method with Saliency Analysis and Joint Optimization' 的科研主题。它们共同构成独一无二的学术指纹。

引用此