Year: 2021
Author: Chao Ma, Lexing Ying
Journal of Computational Mathematics, Vol. 39 (2021), Iss. 6 : pp. 880–896
Abstract
A new understanding of adversarial examples and adversarial robustness is proposed by decoupling the data generator and the label generator (which we call the teacher). In our framework, adversarial robustness is a conditional concept — the student model is not absolutely robust, but robust with respect to the teacher. Based on the new understanding, we claim that adversarial examples exist because the student cannot obtain sufficient information of the teacher from the training data. Various ways of achieving robustness is compared. Theoretical and numerical evidence shows that to efficiently attain robustness, a teacher that actively provides its information to the student may be necessary.
You do not have full access to this article.
Already a Subscriber? Sign in as an individual or via your institution
Journal Article Details
Publisher Name: Global Science Press
Language: English
DOI: https://doi.org/10.4208/jcm.2105-m2020-0310
Journal of Computational Mathematics, Vol. 39 (2021), Iss. 6 : pp. 880–896
Published online: 2021-01
AMS Subject Headings:
Copyright: COPYRIGHT: © Global Science Press
Pages: 17
Keywords: Adversarial robustness Decoupled supervised learning Active teacher.
Author Details
-
LDN-RC: a lightweight denoising network with residual connection to improve adversarial robustness
Chai, Xiuli
Wei, Tongtong
Chen, Zhen
He, Xin
Gan, Zhihua
Wu, Xiangjun
(2022)
https://doi.org/10.1007/s10489-022-03847-z [Citations: 1]