Saved in:
Bibliographic Details
Main Authors: Chen, Wenlong, Klochkov, Yegor, Liu, Yang
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2310.05725
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916159940460544
author Chen, Wenlong
Klochkov, Yegor
Liu, Yang
author_facet Chen, Wenlong
Klochkov, Yegor
Liu, Yang
contents We consider a binary classification problem under group fairness constraints, which can be one of Demographic Parity (DP), Equalized Opportunity (EOp), or Equalized Odds (EO). We propose an explicit characterization of Bayes optimal classifier under the fairness constraints, which turns out to be a simple modification rule of the unconstrained classifier. Namely, we introduce a novel instance-level measure of bias, which we call bias score, and the modification rule is a simple linear rule on top of the finite amount of bias scores.Based on this characterization, we develop a post-hoc approach that allows us to adapt to fairness constraints while maintaining high accuracy. In the case of DP and EOp constraints, the modification rule is thresholding a single bias score, while in the case of EO constraints we are required to fit a linear modification rule with 2 parameters. The method can also be applied for composite group-fairness criteria, such as ones involving several sensitive attributes.
format Preprint
id arxiv_https___arxiv_org_abs_2310_05725
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Post-hoc Bias Scoring Is Optimal For Fair Classification
Chen, Wenlong
Klochkov, Yegor
Liu, Yang
Machine Learning
We consider a binary classification problem under group fairness constraints, which can be one of Demographic Parity (DP), Equalized Opportunity (EOp), or Equalized Odds (EO). We propose an explicit characterization of Bayes optimal classifier under the fairness constraints, which turns out to be a simple modification rule of the unconstrained classifier. Namely, we introduce a novel instance-level measure of bias, which we call bias score, and the modification rule is a simple linear rule on top of the finite amount of bias scores.Based on this characterization, we develop a post-hoc approach that allows us to adapt to fairness constraints while maintaining high accuracy. In the case of DP and EOp constraints, the modification rule is thresholding a single bias score, while in the case of EO constraints we are required to fit a linear modification rule with 2 parameters. The method can also be applied for composite group-fairness criteria, such as ones involving several sensitive attributes.
title Post-hoc Bias Scoring Is Optimal For Fair Classification
topic Machine Learning
url https://arxiv.org/abs/2310.05725