TY - GEN
T1 - HEPT Attack
T2 - 32nd ACM International Conference on Information and Knowledge Management, CIKM 2023
AU - Li, Qi
AU - Li, Xingyu
AU - Cui, Xiaodong
AU - Tang, Keke
AU - Zhu, Peican
N1 - Publisher Copyright:
© 2023 Copyright held by the owner/author(s). Publication rights licensed to ACM.
PY - 2023/10/21
Y1 - 2023/10/21
N2 - Exploring adversarial attacks on deep neural networks (DNNs) is crucial for assessing and enhancing their adversarial robustness. Among various attack types, hard-label attacks that rely only on predicted labels offer a practical approach. This paper focuses on the challenging task of hard-label attacks within an extremely limited query budget, which is a significant achievement rarely accomplished by existing methods. To tackle this, we propose an attack framework that leverages geometric information from previous perturbation directions to form triangles and employs a heuristic perpendicular trial to effectively utilize the intermediate directions. Extensive experiments validate the effectiveness of our approach under strict query constraints and demonstrate its superiority to the state-of-the-art methods.
AB - Exploring adversarial attacks on deep neural networks (DNNs) is crucial for assessing and enhancing their adversarial robustness. Among various attack types, hard-label attacks that rely only on predicted labels offer a practical approach. This paper focuses on the challenging task of hard-label attacks within an extremely limited query budget, which is a significant achievement rarely accomplished by existing methods. To tackle this, we propose an attack framework that leverages geometric information from previous perturbation directions to form triangles and employs a heuristic perpendicular trial to effectively utilize the intermediate directions. Extensive experiments validate the effectiveness of our approach under strict query constraints and demonstrate its superiority to the state-of-the-art methods.
KW - deep neural networks
KW - hard-label adversarial attack
KW - trustworthy machine learning
UR - https://www.scopus.com/pages/publications/85178098776
U2 - 10.1145/3583780.3615198
DO - 10.1145/3583780.3615198
M3 - 会议稿件
AN - SCOPUS:85178098776
T3 - International Conference on Information and Knowledge Management, Proceedings
SP - 4064
EP - 4068
BT - CIKM 2023 - Proceedings of the 32nd ACM International Conference on Information and Knowledge Management
PB - Association for Computing Machinery
Y2 - 21 October 2023 through 25 October 2023
ER -