Fair Classification with Group-Dependent Label Noise

TL;DR

This paper investigates training fair classifiers under group-dependent label noise, revealing that naive fairness constraints can harm both accuracy and fairness, and proposes surrogate loss methods to mitigate these issues.

Contribution

It introduces a novel approach using surrogate loss functions to improve fair classification in the presence of heterogeneous label noise, with theoretical and empirical validation.

Findings

Naive fairness constraints can decrease accuracy and fairness.

Surrogate loss functions improve fairness under label noise.

Methods are validated both theoretically and empirically.

Abstract

This work examines how to train fair classifiers in settings where training labels are corrupted with random noise, and where the error rates of corruption depend both on the label class and on the membership function for a protected subgroup. Heterogeneous label noise models systematic biases towards particular groups when generating annotations. We begin by presenting analytical results which show that naively imposing parity constraints on demographic disparity measures, without accounting for heterogeneous and group-dependent error rates, can decrease both the accuracy and the fairness of the resulting classifier. Our experiments demonstrate these issues arise in practice as well. We address these problems by performing empirical risk minimization with carefully defined surrogate loss functions and surrogate constraints that help avoid the pitfalls introduced by heterogeneous label noise. We provide both theoretical and empirical justifications for the efficacy of our methods. We view our results as an important example of how imposing fairness on biased data sets without proper care can do at least as much harm as it does good.