Towards A Scalable Solution for Improving Multi-Group Fairness in Compositional Classification

TL;DR

This paper addresses the challenge of improving fairness in complex multi-group, multi-label classification systems by proposing scalable techniques that outperform baseline methods in efficiency and effectiveness.

Contribution

Introduces task-overconditioning and group-interleaving methods that enable constant scaling for fairness remediation in multi-group, multi-label classification systems.

Findings

Baseline approaches scale linearly with groups and labels, becoming impractical.

Proposed techniques achieve constant scaling in multi-group multi-label fairness.

Experimental results show effective mitigation in academic and real-world environments.

Abstract

Despite the rich literature on machine learning fairness, relatively little attention has been paid to remediating complex systems, where the final prediction is the combination of multiple classifiers and where multiple groups are present. In this paper, we first show that natural baseline approaches for improving equal opportunity fairness scale linearly with the product of the number of remediated groups and the number of remediated prediction labels, rendering them impractical. We then introduce two simple techniques, called {\em task-overconditioning} and {\em group-interleaving}, to achieve a constant scaling in this multi-group multi-label setup. Our experimental results in academic and real-world environments demonstrate the effectiveness of our proposal at mitigation within this environment.