Hashing Fake: Producing Adversarial Perturbation for Online Privacy Protection Against Automatic Retrieval Models

doi:10.1109/TCSS.2022.3204120

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 互联网大数据与信息安全

	Hashing Fake: Producing Adversarial Perturbation for Online Privacy Protection Against Automatic Retrieval Models
	Zhang, Xingwei 1,2; Zheng, Xiaolong1,2 ; Mao, Wenji1,2 ; Zeng, Daniel Dajun1,2 ; Wang, Fei-Yue1,2
发表期刊	IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
ISSN	2329-924X
	2022-09-30
页码	11
通讯作者	Zheng, Xiaolong(xiaolong.zheng@ia.ac.cn)
摘要	The wide application of deep neural networks (DNNs) has significantly improved the performance of hashing models on multimodal retrieval issues. DNN-based deep models can automatically learn semantic features from raw data to make human-level decisions. However, the superior generalization leads to potential privacy leakage risks. Strong DNN-based retrieval models enable malicious crawlers to search for nontag private information based on semantic similarity matching. Hence, executing effective privacy protection mechanisms against those retrieval software is essential for reliable social website construction. In this article, we propose a retrieval task-based adversarial perturbation generation method called Hashing Fake to meet this request. Specifically, DNNs are recently found to be vulnerable to a specific set of attacks called adversarial perturbations, which denote some magnitude-restricted signals added on objective samples to misguide well-crafted DNN models, and perturbations' magnitudes are small enough that will not induce humans' perception. Moreover, since existing adversarial perturbation generation methods are designed for supervised tasks, Hashing Fake constructs a differential approximation substitution for perturbation production on unsupervised retrieval tasks. Through extensive experiments on several deep retrieval benchmarks, we demonstrate that well-crafted perturbations using Hashing Fake can effectively misguide objective models' recognitions to make false predictions. The added norm-restricted perturbations on objective samples will not alter humans' perception; hence, Hashing Fake can be applied on real-world social websites to protect subscribers' privacy against malicious retrieval software.
关键词	Perturbation methods Semantics Computational modeling Codes Task analysis Privacy Software Adversarial perturbation deep neural network (DNN) privacy protection semantic retrieval
DOI	10.1109/TCSS.2022.3204120
收录类别	SCI
语种	英语
资助项目	Ministry of Science and Technology of China[2020AAA0108401] ; Natural Science Foundation of China[72225011] ; Natural Science Foundation of China[71621002]
项目资助者	Ministry of Science and Technology of China ; Natural Science Foundation of China
WOS研究方向	Computer Science
WOS类目	Computer Science, Cybernetics ; Computer Science, Information Systems
WOS记录号	WOS:000865074900001
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/50353
专题	多模态人工智能系统全国重点实验室_互联网大数据与信息安全多模态人工智能系统全国重点实验室_平行智能技术与系统团队
通讯作者	Zheng, Xiaolong
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 101408, Peoples R China 2.Chinese Acad Sci, State Key Lab Management & Control Complex Syst, Inst Automat, Beijing 100080, Peoples R China
第一作者单位	中国科学院自动化研究所
通讯作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Zhang, Xingwei,Zheng, Xiaolong,Mao, Wenji,et al. Hashing Fake: Producing Adversarial Perturbation for Online Privacy Protection Against Automatic Retrieval Models[J]. IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS,2022:11.
APA	Zhang, Xingwei,Zheng, Xiaolong,Mao, Wenji,Zeng, Daniel Dajun,&Wang, Fei-Yue.(2022).Hashing Fake: Producing Adversarial Perturbation for Online Privacy Protection Against Automatic Retrieval Models.IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS,11.
MLA	Zhang, Xingwei,et al."Hashing Fake: Producing Adversarial Perturbation for Online Privacy Protection Against Automatic Retrieval Models".IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS (2022):11.