Towards a Flexible Framework for Algorithmic Fairness

TL;DR

This paper introduces a flexible algorithmic fairness framework using optimal transport, aiming to adapt to diverse legal and normative fairness requirements while highlighting ongoing normative and legal challenges.

Contribution

It presents a novel optimal transport-based algorithm that interpolates between various fairness definitions, addressing the need for adaptable fairness frameworks in AI.

Findings

The framework can interpolate between multiple fairness definitions.

Legal and normative challenges remain for real-world implementation.

The approach aligns with EU legal standards on algorithmic discrimination.

Abstract

Increasingly, scholars seek to integrate legal and technological insights to combat bias in AI systems. In recent years, many different definitions for ensuring non-discrimination in algorithmic decision systems have been put forward. In this paper, we first briefly describe the EU law framework covering cases of algorithmic discrimination. Second, we present an algorithm that harnesses optimal transport to provide a flexible framework to interpolate between different fairness definitions. Third, we show that important normative and legal challenges remain for the implementation of algorithmic fairness interventions in real-world scenarios. Overall, the paper seeks to contribute to the quest for flexible technical frameworks that can be adapted to varying legal and normative fairness constraints.