Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatment

TL;DR

This paper introduces a new fairness notion called disparate mistreatment, focusing on equalizing misclassification rates across social groups, and proposes methods to incorporate this fairness into classifiers with minimal accuracy loss.

Contribution

It defines disparate mistreatment as a fairness criterion based on misclassification rates and develops convex constraints to enforce this fairness in decision boundary classifiers.

Findings

Effectively reduces disparate mistreatment in classifiers

Maintains high accuracy with minimal trade-offs

Works on both synthetic and real datasets

Abstract

Automated data-driven decision making systems are increasingly being used to assist, or even replace humans in many settings. These systems function by learning from historical decisions, often taken by humans. In order to maximize the utility of these systems (or, classifiers), their training involves minimizing the errors (or, misclassifications) over the given historical data. However, it is quite possible that the optimally trained classifier makes decisions for people belonging to different social groups with different misclassification rates (e.g., misclassification rates for females are higher than for males), thereby placing these groups at an unfair disadvantage. To account for and avoid such unfairness, in this paper, we introduce a new notion of unfairness, disparate mistreatment, which is defined in terms of misclassification rates. We then propose intuitive measures of disparate mistreatment for decision boundary-based classifiers, which can be easily incorporated into their formulation as convex-concave constraints. Experiments on synthetic as well as real world datasets show that our methodology is effective at avoiding disparate mistreatment, often at a small cost in terms of accuracy.