Projection to Fairness in Statistical Learning
Machine Learning
2020-06-26 v4 Statistics Theory
Machine Learning
Statistics Theory
Abstract
In the context of regression, we consider the fundamental question of making an estimator fair while preserving its prediction accuracy as much as possible. To that end, we define its projection to fairness as its closest fair estimator in a sense that reflects prediction accuracy. Our methodology leverages tools from optimal transport to construct efficiently the projection to fairness of any given estimator as a simple post-processing step. Moreover, our approach precisely quantifies the cost of fairness, measured in terms of prediction accuracy.
Cite
@article{arxiv.2005.11720,
title = {Projection to Fairness in Statistical Learning},
author = {Thibaut Le Gouic and Jean-Michel Loubes and Philippe Rigollet},
journal= {arXiv preprint arXiv:2005.11720},
year = {2020}
}