Fairness Explainability using Optimal Transport with Applications in Image Classification

TL;DR

This paper introduces a novel fairness explainability method using optimal transport theory, specifically Wasserstein barycenters, to identify and mitigate biases in image classification models, enhancing transparency and accountability.

Contribution

It presents a new approach that combines fairness enforcement with explainability using optimal transport, addressing the gap in understanding why models are biased.

Findings

Uses Wasserstein barycenters to achieve fair predictions

Introduces an extension to identify bias-related regions

Demonstrates improved fairness and interpretability in image classification

Abstract

Ensuring trust and accountability in Artificial Intelligence systems demands explainability of its outcomes. Despite significant progress in Explainable AI, human biases still taint a substantial portion of its training data, raising concerns about unfairness or discriminatory tendencies. Current approaches in the field of Algorithmic Fairness focus on mitigating such biases in the outcomes of a model, but few attempts have been made to try to explain \emph{why} a model is biased. To bridge this gap between the two fields, we propose a comprehensive approach that uses optimal transport theory to uncover the causes of discrimination in Machine Learning applications, with a particular emphasis on image classification. We leverage Wasserstein barycenters to achieve fair predictions and introduce an extension to pinpoint bias-associated regions. This allows us to derive a cohesive system which uses the enforced fairness to measure each features influence \emph{on} the bias. Taking advantage of this interplay of enforcing and explaining fairness, our method hold significant implications for the development of trustworthy and unbiased AI systems, fostering transparency, accountability, and fairness in critical decision-making scenarios across diverse domains.