Adaptive Perturbation for Adversarial Attack

doi:10.1109/TPAMI.2024.3367773

	Adaptive Perturbation for Adversarial Attack
	Yuan, Zheng 1,2; Zhang, Jie 1,2; Jiang, Zhaoyan 3; Li, Liangliang ; Shan, Shiguang 2
	2024-08-01
发表期刊	IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
ISSN	0162-8828
卷号	46 期号:8 页码:5663-5676
摘要	In recent years, the security of deep learning models achieves more and more attentions with the rapid development of neural networks, which are vulnerable to adversarial examples. Almost all existing gradient-based attack methods use the sign function in the generation to meet the requirement of perturbation budget on L-infinity norm. However, we find that the sign function may be improper for generating adversarial examples since it modifies the exact gradient direction. Instead of using the sign function, we propose to directly utilize the exact gradient direction with a scaling factor for generating adversarial perturbations, which improves the attack success rates of adversarial examples even with fewer perturbations. At the same time, we also theoretically prove that this method can achieve better black-box transferability. Moreover, considering that the best scaling factor varies across different images, we propose an adaptive scaling factor generator to seek an appropriate scaling factor for each image, which avoids the computational cost for manually searching the scaling factor. Our method can be integrated with almost all existing gradient-based attack methods to further improve their attack success rates. Extensive experiments on the CIFAR10 and ImageNet datasets show that our method exhibits higher transferability and outperforms the state-of-the-art methods.
关键词	Perturbation methods Iterative methods Adaptation models Generators Closed box Security Training Adversarial attack transfer-based attack adversarial example adaptive perturbation
DOI	10.1109/TPAMI.2024.3367773
收录类别	SCI
语种	英语
资助项目	National Key R&D Program of China[2021YFC3310100] ; National Natural Science Foundation of China[62176251] ; Beijing Nova Program[20230484368] ; Youth Innovation Promotion Association CAS
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:001262841000014
出版者	IEEE COMPUTER SOC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/39845
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Zhang, Jie
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100049, Peoples R China 3.Tencent, Shenzhen 518057, Peoples R China
推荐引用方式 GB/T 7714	Yuan, Zheng,Zhang, Jie,Jiang, Zhaoyan,et al. Adaptive Perturbation for Adversarial Attack[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2024,46(8):5663-5676.
APA	Yuan, Zheng,Zhang, Jie,Jiang, Zhaoyan,Li, Liangliang,&Shan, Shiguang.(2024).Adaptive Perturbation for Adversarial Attack.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,46(8),5663-5676.
MLA	Yuan, Zheng,et al."Adaptive Perturbation for Adversarial Attack".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 46.8(2024):5663-5676.