To be Robust or to be Fair: Towards Fairness in Adversarial Training

Authors: Han Xu, Xiaorui Liu, Yaxin Li, Anil K. Jain, Jiliang Tang
Published: 2020-10-13 | Updated: 2021-05-18

Source: https://arxiv.org/abs/2010.06121

PDF: https://arxiv.org/pdf/2010.06121

Labels Predicted by AI

敵対的訓練 公平性の確保 バイアス緩和手法

Please note that these labels were automatically added by AI. Therefore, they may not be entirely accurate.
For more details, please see the About the Literature Database page.

Abstract

Adversarial training algorithms have been proved to be reliable to improve machine learning models’ robustness against adversarial examples. However, we find that adversarial training algorithms tend to introduce severe disparity of accuracy and robustness between different groups of data. For instance, a PGD adversarially trained ResNet18 model on CIFAR-10 has 93 PGD l-infty-8 robust accuracy on the class “automobile” but only 65 the class “cat”. This phenomenon happens in balanced datasets and does not exist in naturally trained models when only using clean samples. In this work, we empirically and theoretically show that this phenomenon can happen under general adversarial training algorithms which minimize DNN models’ robust errors. Motivated by these findings, we propose a Fair-Robust-Learning (FRL) framework to mitigate this unfairness problem when doing adversarial defenses. Experimental results validate the effectiveness of FRL.