Boosting Black-Box Adversarial Attacks with Meta Learning

Junjie Fu, Jian Sun, Gang Wang

科研成果: 书/报告/会议事项章节会议稿件同行评审

3 引用 (Scopus)

摘要

Deep neural networks (DNNs) have achieved remarkable success in diverse fields. However, it has been demonstrated that DNNs are very vulnerable to adversarial examples even in black-box settings. A large number of black-box attack methods have been proposed to in the literature. However, those methods usually suffer from low success rates and large query counts, which cannot fully satisfy practical purposes. In this paper, we propose a hybrid attack method which trains meta adversarial perturbations (MAPs) on surrogate models and performs black-box attacks by estimating gradients of the models. Our method uses the meta adversarial perturbation as an initialization and subsequently trains any black-box attack method for several epochs. Furthermore, the MAPs enjoy favorable transferability and universality, in the sense that they can be employed to boost performance of other black-box adversarial attack methods. Extensive experiments demonstrate that our method can not only improve the attack success rates, but also reduces the number of queries compared to other methods.

源语言英语
主期刊名Proceedings of the 41st Chinese Control Conference, CCC 2022
编辑Zhijun Li, Jian Sun
出版商IEEE Computer Society
7308-7313
页数6
ISBN(电子版)9789887581536
DOI
出版状态已出版 - 2022
活动41st Chinese Control Conference, CCC 2022 - Hefei, 中国
期限: 25 7月 202227 7月 2022

出版系列

姓名Chinese Control Conference, CCC
2022-July
ISSN(印刷版)1934-1768
ISSN(电子版)2161-2927

会议

会议41st Chinese Control Conference, CCC 2022
国家/地区中国
Hefei
时期25/07/2227/07/22

指纹

探究 'Boosting Black-Box Adversarial Attacks with Meta Learning' 的科研主题。它们共同构成独一无二的指纹。

引用此