A Feature Guided Denoising Network For Adversarial Defense

Jinhui Li, Dahao Xu, Yining Qin, Xinyang Deng

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

As neural networks are playing a more and more significant role in many fields, they are also under the risk of being attacked by adversarial examples, which becomes a great challenge to the whole community. Due to this problem, networks cannot provide the guaranteed security when they are used in some security-sensitive scenarios. In this paper, a feature guided denoising network for adversarial defense is studied. Specifically, a denoising network is designed to remove possible adversarial perturbations on the input image. and a novel training method of the denoising network is proposed to improve the performance, in which deep features extracted from clean examples by the pretrained classifier is used as supervision information in the training process. Experimental results reveal that the proposed method shows satisfactory performance on defending against several white-box adversarial attacks. Besides, combination of the proposed method and adversarial training is studied, which achieves very good results compared to the other experiments reported in this paper.

源语言英语
主期刊名Proceedings of 2022 IEEE International Conference on Unmanned Systems, ICUS 2022
编辑Rong Song
出版商Institute of Electrical and Electronics Engineers Inc.
393-398
页数6
ISBN(电子版)9781665484565
DOI
出版状态已出版 - 2022
活动2022 IEEE International Conference on Unmanned Systems, ICUS 2022 - Guangzhou, 中国
期限: 28 10月 202230 10月 2022

出版系列

姓名Proceedings of 2022 IEEE International Conference on Unmanned Systems, ICUS 2022

会议

会议2022 IEEE International Conference on Unmanned Systems, ICUS 2022
国家/地区中国
Guangzhou
时期28/10/2230/10/22

指纹

探究 'A Feature Guided Denoising Network For Adversarial Defense' 的科研主题。它们共同构成独一无二的指纹。

引用此