Learning Fair Representations with High-Confidence Guarantees

TL;DR

This paper introduces a new framework called FRG for learning fair data representations with high-confidence guarantees, ensuring fairness across multiple tasks and models.

Contribution

The paper formally defines high-confidence fairness guarantees in representation learning and proposes the FRG framework to achieve this across all downstream tasks.

Findings

FRG provides high-confidence fairness guarantees for multiple tasks.

Empirical results show FRG effectively limits unfairness in downstream models.

FRG outperforms baseline methods in fairness metrics.

Abstract

Representation learning is increasingly employed to generate representations that are predictive across multiple downstream tasks. The development of representation learning algorithms that provide strong fairness guarantees is thus important because it can prevent unfairness towards disadvantaged groups for all downstream prediction tasks. To prevent unfairness towards disadvantaged groups in all downstream tasks, it is crucial to provide representation learning algorithms that provide fairness guarantees. In this paper, we formally define the problem of learning representations that are fair with high confidence. We then introduce the Fair Representation learning with high-confidence Guarantees (FRG) framework, which provides high-confidence guarantees for limiting unfairness across all downstream models and tasks, with user-defined upper bounds. After proving that FRG ensures fairness for all downstream models and tasks with high probability, we present empirical evaluations that demonstrate FRG's effectiveness at upper bounding unfairness for multiple downstream models and tasks.