Detecting Adversarial Examples Utilizing Pixel Value Diversity

doi:10.1109/AsianHOST53231.2021.9699495

	Detecting Adversarial Examples Utilizing Pixel Value Diversity
	Dong, Jinxin; Zhou, Pingqiang
	2021-12
会议录名称	2021 ASIAN HARDWARE ORIENTED SECURITY AND TRUST SYMPOSIUM (ASIANHOST)
发表状态	已发表
DOI	10.1109/AsianHOST53231.2021.9699495
摘要	The state-of-the-art deep learning methods can be vulnerable: imperceptibly elaborated perturbations will induce unexpected behaviors. In this paper, we introduce two novel adversarial example detection methods utilizing pixel value diversity. First, we propose two independent metrics to assess the pixel value diversity separately, which reflects the spread of the pixel values in an image. Then we observe that adversarial examples are different from clean images on both metrics, regardless of attack methods. Based on this observation, for either metric, we can set a threshold and compare the threshold with the value of an image on the metric to detect whether the image is an adversarial example. Against several popular attack methods, experimental results on a variety of datasets show that our approach achieves better performances in detecting adversarial examples, compared to the state-of-the-art detection method. We also show that our methods are reliable even against adaptive attack.
会议录编者/会议主办者	IEEE,IEEE Council Elect Design Automat,IEEE Hardware Secur & Trust Tech Comm,China Comp Federat,ShanghaiTech Univ,Chinese Acad Sci, Inst Comp Technol
会议名称	IEEE Asian Hardware Oriented Security and Trust Symposium (AsianHOST)
出版地	345 E 47TH ST, NEW YORK, NY 10017 USA
会议地点	null,Shanghai,PEOPLES R CHINA
会议日期	DEC 16-18, 2021
URL	查看原文
收录类别	CPCI-S ; EI
语种	英语
WOS研究方向	Computer Science
WOS类目	Computer Science, Hardware & Architecture ; Computer Science, Information Systems
WOS记录号	WOS:000799493300003
出版者	IEEE
EI入藏号	978-1-6654-4185-8
原始文献类型	Proceedings Paper
来源库	IEEE
引用统计	正在获取...
文献类型	会议论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/155926
专题	信息科学与技术学院_硕士生信息科学与技术学院_PI研究组_周平强组
通讯作者	Dong, Jinxin
作者单位	ShanghaiTech Univ, Sch Informat Sci & Technol, Shanghai, Peoples R China
第一作者单位	信息科学与技术学院
通讯作者单位	信息科学与技术学院
第一作者的第一单位	信息科学与技术学院
推荐引用方式 GB/T 7714	Dong, Jinxin,Zhou, Pingqiang. Detecting Adversarial Examples Utilizing Pixel Value Diversity[C]//IEEE,IEEE Council Elect Design Automat,IEEE Hardware Secur & Trust Tech Comm,China Comp Federat,ShanghaiTech Univ,Chinese Acad Sci, Inst Comp Technol. 345 E 47TH ST, NEW YORK, NY 10017 USA:IEEE,2021.