Moving target defense against adversarial attacks
Deep neural network has been successfully applied to image classification, but recent research work shows that deep neural network is vulnerable to adversarial attacks.A moving target defense method was proposed by means of dynamic switching model with a Bayes-Stackelberg game strategy, which could...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
POSTS&TELECOM PRESS Co., LTD
2021-02-01
|
| Series: | 网络与信息安全学报 |
| Subjects: | |
| Online Access: | http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2021012 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841529862979846144 |
|---|---|
| author | Bin WANG Liang CHEN Yaguan QIAN Yankai GUO Qiqi SHAO Jiamin WANG |
| author_facet | Bin WANG Liang CHEN Yaguan QIAN Yankai GUO Qiqi SHAO Jiamin WANG |
| author_sort | Bin WANG |
| collection | DOAJ |
| description | Deep neural network has been successfully applied to image classification, but recent research work shows that deep neural network is vulnerable to adversarial attacks.A moving target defense method was proposed by means of dynamic switching model with a Bayes-Stackelberg game strategy, which could prevent an attacker from continuously obtaining consistent information and thus blocked its construction of adversarial examples.To improve the defense effect of the proposed method, the gradient consistency among the member models was taken as a measure to construct a new loss function in training for improving the difference among the member models.Experimental results show that the proposed method can improve the moving target defense performance of the image classification system and significantly reduce the attack success rate against the adversarial examples. |
| format | Article |
| id | doaj-art-2ad86d8269f342ecbf0370deb2daed6f |
| institution | Kabale University |
| issn | 2096-109X |
| language | English |
| publishDate | 2021-02-01 |
| publisher | POSTS&TELECOM PRESS Co., LTD |
| record_format | Article |
| series | 网络与信息安全学报 |
| spelling | doaj-art-2ad86d8269f342ecbf0370deb2daed6f2025-01-15T03:14:42ZengPOSTS&TELECOM PRESS Co., LTD网络与信息安全学报2096-109X2021-02-01711312059562993Moving target defense against adversarial attacksBin WANGLiang CHENYaguan QIANYankai GUOQiqi SHAOJiamin WANGDeep neural network has been successfully applied to image classification, but recent research work shows that deep neural network is vulnerable to adversarial attacks.A moving target defense method was proposed by means of dynamic switching model with a Bayes-Stackelberg game strategy, which could prevent an attacker from continuously obtaining consistent information and thus blocked its construction of adversarial examples.To improve the defense effect of the proposed method, the gradient consistency among the member models was taken as a measure to construct a new loss function in training for improving the difference among the member models.Experimental results show that the proposed method can improve the moving target defense performance of the image classification system and significantly reduce the attack success rate against the adversarial examples.http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2021012adversarial examplesmoving target defenseBayes-Stackelberg game |
| spellingShingle | Bin WANG Liang CHEN Yaguan QIAN Yankai GUO Qiqi SHAO Jiamin WANG Moving target defense against adversarial attacks 网络与信息安全学报 adversarial examples moving target defense Bayes-Stackelberg game |
| title | Moving target defense against adversarial attacks |
| title_full | Moving target defense against adversarial attacks |
| title_fullStr | Moving target defense against adversarial attacks |
| title_full_unstemmed | Moving target defense against adversarial attacks |
| title_short | Moving target defense against adversarial attacks |
| title_sort | moving target defense against adversarial attacks |
| topic | adversarial examples moving target defense Bayes-Stackelberg game |
| url | http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2021012 |
| work_keys_str_mv | AT binwang movingtargetdefenseagainstadversarialattacks AT liangchen movingtargetdefenseagainstadversarialattacks AT yaguanqian movingtargetdefenseagainstadversarialattacks AT yankaiguo movingtargetdefenseagainstadversarialattacks AT qiqishao movingtargetdefenseagainstadversarialattacks AT jiaminwang movingtargetdefenseagainstadversarialattacks |