Fairness-informed Pareto Optimization : An Efficient Bilevel Framework

TL;DR

This paper introduces BADR, a bilevel framework that efficiently finds Pareto-efficient models for any fairness metric in machine learning, addressing limitations of existing approaches.

Contribution

The paper proposes a novel bilevel adaptive rescaling framework with two algorithms, BADR-GD and BADR-SGD, providing convergence guarantees and broad fairness metric applicability.

Findings

BADR outperforms existing Pareto-efficient fairness methods in experiments.

The framework is versatile across various fairness metrics and learning tasks.

Open-source toolbox 'badr' facilitates practical adoption.

Abstract

Despite their promise, fair machine learning methods often yield Pareto-inefficient models, in which the performance of certain groups can be improved without degrading that of others. This issue arises frequently in traditional in-processing approaches such as fairness-through-regularization. In contrast, existing Pareto-efficient approaches are biased towards a certain perspective on fairness and fail to adapt to the broad range of fairness metrics studied in the literature. In this paper, we present BADR, a simple framework to recover the optimal Pareto-efficient model for any fairness metric. Our framework recovers its models through a Bilevel Adaptive Rescalarisation procedure. The lower level is a weighted empirical risk minimization task where the weights are a convex combination of the groups, while the upper level optimizes the chosen fairness objective. We equip our framework with two novel large-scale, single-loop algorithms, BADR-GD and BADR-SGD, and establish their convergence guarantees. We release badr, an open-source Python toolbox implementing our framework for a variety of learning tasks and fairness metrics. Finally, we conduct extensive numerical experiments demonstrating the advantages of BADR over existing Pareto-efficient approaches to fairness.