TY - JOUR T1 - Achieving Adversarial Robustness Requires an Active Teacher AU - Ma , Chao AU - Ying , Lexing JO - Journal of Computational Mathematics VL - 6 SP - 880 EP - 896 PY - 2021 DA - 2021/10 SN - 39 DO - http://doi.org/10.4208/jcm.2105-m2020-0310 UR - https://global-sci.org/intro/article_detail/jcm/19916.html KW - Adversarial robustness, Decoupled supervised learning, Active teacher. AB -

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.