Confidence Intervals for Testing Disparate Impact in Fair Learning

Jean-Michel Loubes, Philippe Besse, Eustasio del Barrio, Paula Gordaliza

Publication date: 17 July 2018

Abstract: We provide the asymptotic distribution of the major indexes used in the statistical literature to quantify disparate treatment in machine learning. We aim at promoting the use of confidence intervals when testing the so-called group disparate impact. We illustrate on some examples the importance of using confidence intervals and not a single value.

Has companion code repository: https://github.com/JMLToulouse/FairLearning

This page was built for publication: Confidence Intervals for Testing Disparate Impact in Fair Learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6304307)