Implementing Fair Regression In The Real World

TL;DR

This paper examines how fair regression algorithms affect individual predictions and proposes post-processing methods to enhance their utility, addressing the gap between group fairness and individual impact.

Contribution

It investigates the individual-level effects of fair regression and introduces post-processing algorithms to improve their practical utility.

Findings

Fair regression alters individual predictions compared to baseline.

Post-processing algorithms can enhance the utility of fair regression.

The study provides insights into balancing fairness and individual accuracy.

Abstract

Most fair regression algorithms mitigate bias towards sensitive sub populations and therefore improve fairness at group level. In this paper, we investigate the impact of such implementation of fair regression on the individual. More precisely, we assess the evolution of continuous predictions from an unconstrained to a fair algorithm by comparing results from baseline algorithms with fair regression algorithms for the same data points. Based on our findings, we propose a set of post-processing algorithms to improve the utility of the existing fair regression approaches.