MultiPAD: A Multivariant Partition-Based Method for Audio Adversarial Examples Detection

doi:10.1109/ACCESS.2020.2985231

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	MultiPAD: A Multivariant Partition-Based Method for Audio Adversarial Examples Detection
	Guo, Qingli 2,3; Ye, Jing 2,3; Hu, Yu 2,3; Zhang, Guohe 1; Li, Xiaowei 2,3; Li, Huawei 2,3,4
刊名	IEEE ACCESS
	2020
卷号	8 页码:63368-63380
关键词	Speech recognition Feature extraction Decoding Mathematical model Acoustics Psychoacoustic models Radio frequency Adversarial examples audio detection multivariant partition
ISSN号	2169-3536
DOI	10.1109/ACCESS.2020.2985231
英文摘要	Adversarial examples have been highlighted as a serious threat to various deep neural networks. The defense against adversarial examples is extremely urgent. This paper proposes an efficient multivariant partition based method to detect audio adversarial examples. Various partition strategies are exploited to obtain sufficient features that can help us to distinguish audio adversarial examples from clean samples. Using these features, a classification model is trained to detect audio adversarial examples. These features are also combined and compared to analyze their detection performance. The performance is evaluated on the Mozilla Common Voice dataset and the LibriSpeech dataset. Experimental results based on Mozilla Common Voice dataset show that the detection accuracy and AUC value of the model achieve 94.8 & x0025; and 0.97 respectively, which are 13.5 & x0025; and 0.08 higher than using the features of the existing work. Experimental results based on LibriSpeech dataset show that the detection accuracy and AUC value of the model achieve 100 & x0025; and 1.00 respectively, which are 10 & x0025; and 0.10 higher than the existing work.
资助项目	National Natural Science Foundation of China (NSFC)[61532017] ; National Natural Science Foundation of China (NSFC)[61704174] ; National Natural Science Foundation of China (NSFC)[61432017] ; National Natural Science Foundation of China (NSFC)[61521092]
WOS研究方向	Computer Science ; Engineering ; Telecommunications
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000530832200063
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/15355]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Ye, Jing; Li, Xiaowei
作者单位	1.Xi An Jiao Tong Univ, Sch Microelect, Xian 710049, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 4.Peng Cheng Lab, Shenzhen 518052, Peoples R China
推荐引用方式 GB/T 7714	Guo, Qingli,Ye, Jing,Hu, Yu,et al. MultiPAD: A Multivariant Partition-Based Method for Audio Adversarial Examples Detection[J]. IEEE ACCESS,2020,8:63368-63380.
APA	Guo, Qingli,Ye, Jing,Hu, Yu,Zhang, Guohe,Li, Xiaowei,&Li, Huawei.(2020).MultiPAD: A Multivariant Partition-Based Method for Audio Adversarial Examples Detection.IEEE ACCESS,8,63368-63380.
MLA	Guo, Qingli,et al."MultiPAD: A Multivariant Partition-Based Method for Audio Adversarial Examples Detection".IEEE ACCESS 8(2020):63368-63380.