Advancing subgroup fairness via sleeping experts

TL;DR

This paper explores methods to improve subgroup fairness in overlapping populations with sequential predictions, revealing challenges and proposing solutions inspired by sleeping experts in online learning, especially under one-sided feedback.

Contribution

It introduces a novel connection between subgroup fairness and sleeping experts, extending results to one-sided feedback scenarios and analyzing incentives within a game-theoretic framework.

Findings

Fairness for overlapping groups can be statistically impossible under certain objectives.

When individuals are equally important to all groups, fairness goals become achievable.

The paper extends online learning techniques to fairness problems with one-sided feedback.

Abstract

We study methods for improving fairness to subgroups in settings with overlapping populations and sequential predictions. Classical notions of fairness focus on the balance of some property across different populations. However, in many applications the goal of the different groups is not to be predicted equally but rather to be predicted well. We demonstrate that the task of satisfying this guarantee for multiple overlapping groups is not straightforward and show that for the simple objective of unweighted average of false negative and false positive rate, satisfying this for overlapping populations can be statistically impossible even when we are provided predictors that perform well separately on each subgroup. On the positive side, we show that when individuals are equally important to the different groups they belong to, this goal is achievable; to do so, we draw a connection to the sleeping experts literature in online learning. Motivated by the one-sided feedback in natural settings of interest, we extend our results to such a feedback model. We also provide a game-theoretic interpretation of our results, examining the incentives of participants to join the system and to provide the system full information about predictors they may possess. We end with several interesting open problems concerning the strength of guarantees that can be achieved in a computationally efficient manner.