Back to Search Start Over

Breaking Fair Binary Classification with Optimal Flipping Attacks

Authors :
Jo, Changhun
Sohn, Jy-yong
Lee, Kangwook
Publication Year :
2022

Abstract

Minimizing risk with fairness constraints is one of the popular approaches to learning a fair classifier. Recent works showed that this approach yields an unfair classifier if the training set is corrupted. In this work, we study the minimum amount of data corruption required for a successful flipping attack. First, we find lower/upper bounds on this quantity and show that these bounds are tight when the target model is the unique unconstrained risk minimizer. Second, we propose a computationally efficient data poisoning attack algorithm that can compromise the performance of fair learning algorithms.

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2204.05472
Document Type :
Working Paper