Leveraging Labeled and Unlabeled Data for Consistent Fair Binary Classification

06/12/2019
by   Evgenii Chzhen, et al.
0

We study the problem of fair binary classification using the notion of Equal Opportunity. It requires the true positive rate to distribute equally across the sensitive groups. Within this setting we show that the fair optimal classifier is obtained by recalibrating the Bayes classifier by a group-dependent threshold. We provide a constructive expression for the threshold. This result motivates us to devise a plug-in classification procedure based on both unlabeled and labeled datasets. While the latter is used to learn the output conditional probability, the former is used for calibration. The overall procedure can be computed in polynomial time and it is shown to be statistically consistent both in terms of classification error and fairness measure. Finally, we present numerical experiments which indicate that our method is often superior or competitive with the state-of-the-art methods on benchmark datasets.

READ FULL TEXT
research
05/10/2019

Optimal rates for F-score binary classification

We study the minimax settings of binary classification with F-score unde...
research
06/19/2022

Fairness-aware Model-agnostic Positive and Unlabeled Learning

With the increasing application of machine learning in high-stake decisi...
research
06/24/2020

Fairness with Overlapping Groups

In algorithmically fair prediction problems, a standard goal is to ensur...
research
06/02/2018

Binary Classification with Karmic, Threshold-Quasi-Concave Metrics

Complex performance measures, beyond the popular measure of accuracy, ar...
research
10/23/2016

Online Classification with Complex Metrics

We present a framework and analysis of consistent binary classification ...
research
02/01/2021

Binary Classification from Multiple Unlabeled Datasets via Surrogate Set Classification

To cope with high annotation costs, training a classifier only from weak...
research
03/06/2018

A Reductions Approach to Fair Classification

We present a systematic approach for achieving fairness in a binary clas...

Please sign up or login with your details

Forgot password? Click here to reset