HEPT Attack: Heuristic Perpendicular Trial for Hard-label Attacks under Limited Query Budgets
PROCEEDINGS OF THE 32ND ACM INTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, CIKM 2023(2023)
摘要
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.
更多查看译文
关键词
trustworthy machine learning,deep neural networks,hard-label adversarial attack
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要