Intrinsic Fairness-Accuracy Tradeoffs under Equalized Odds

TL;DR

This paper investigates the fundamental tradeoff between fairness and accuracy in machine learning under equalized odds, providing theoretical bounds and empirical validation on real datasets, highlighting inherent limitations due to data disparities.

Contribution

It introduces a new upper bound on classifier accuracy as a function of fairness constraints and data statistics, advancing understanding of fairness-accuracy tradeoffs.

Findings

High accuracy under low bias is fundamentally limited by data disparities.

Theoretical bounds align with empirical results on real datasets.

Existing fair classifiers often approach these theoretical limits.

Abstract

With the growing adoption of machine learning (ML) systems in areas like law enforcement, criminal justice, finance, hiring, and admissions, it is increasingly critical to guarantee the fairness of decisions assisted by ML. In this paper, we study the tradeoff between fairness and accuracy under the statistical notion of equalized odds. We present a new upper bound on the accuracy (that holds for any classifier), as a function of the fairness budget. In addition, our bounds also exhibit dependence on the underlying statistics of the data, labels and the sensitive group attributes. We validate our theoretical upper bounds through empirical analysis on three real-world datasets: COMPAS, Adult, and Law School. Specifically, we compare our upper bound to the tradeoffs that are achieved by various existing fair classifiers in the literature. Our results show that achieving high accuracy subject to a low-bias could be fundamentally limited based on the statistical disparity across the groups.