2022/11/04 by Yulu Jin, Jin, Yulu, Lifeng Lai +1
Computer Science · Environmental Science · Social Sciences · #Adversarial Robustness in Machine Learning #Mercury impact and mitigation studies #Ethics and Social Impacts of AI
paper · pdf · doi:10.48550/arxiv.2211.04449
In this paper, we take a first step towards answering the question of how to design fair machine learning algorithms that are robust to adversarial attacks. Using a minimax framework, we aim to design an adversarially robust fair regression model that achieves optimal performance in the presence of an attacker who is able to add a carefully designed adversarial data point to the dataset or perform a rank-one attack on the dataset. By solving the proposed nonsmooth nonconvex-nonconcave minimax problem, the optimal adversary as well as the robust fairness-aware regression model are obtained. For both synthetic data and real-world datasets, numerical results illustrate that the proposed adversarially robust fair models have better performance on poisoned datasets than other fair machine learning models in both prediction accuracy and group-based fairness measure.