Addressing Fairness in Machine Learning Predictions: Strategic Best-Response Fair Discriminant Removed Algorithm

49 Pages Posted: 13 Jun 2019 Last revised: 24 Jan 2024

See all articles by Hajime Shimao

Hajime Shimao

Santa Fe Institute

Warut Khern-am-nuai

McGill University - Desautels Faculty of Management

Karthik Natarajan Kannan

Purdue University

Maxime C. Cohen

Desautels Faculty of Management, McGill University

Date Written: January 17, 2024

Abstract

Machine learning algorithms have become increasingly common and have affect many aspects of our life. However, because the objective of most of the standard, off-the-shelf machine learning algorithms is to maximize the prediction performance, the results produced by these algorithms could be discriminatory. The discrimination issue has gain the interest from both academic researchers and practitioners to develop machine learning algorithms that are fair. Even then, most such algorithms focus on decreasing the disparity in predictions of successful outcomes. However, these algorithms tend to ignore the strategic behavior of prediction subpopulations, resulting in disparity in the behavior of prediction subjects at equilibrium. One exception is those algorithms that use equalized odds as a fairness criterion which can decrease disparity in behavior. However, they cannot be used in many practical settings. We propose a new class of fair machine learning algorithms that alleviate disparity in prediction results, disparity in behavior of prediction subjects, and does not need to account for the sensitive variable explicitly. Our algorithm also complies with the notion of equal treatment and explainable AI, and can be applied to a wide variety of prediction tasks. We demonstrate the theoretical performance of our algorithm in the asymptotic scenario. In addition, we show the practical performance of the proposed algorithm by comparing its performance with that of other ordinary off-the-shelf algorithms and that of existing fair machine learning algorithms available in the IBM Fairness 360 suite.

Keywords: machine learning, discrimination, fairness, prediction

Suggested Citation

Shimao, Hajime and Khern-am-nuai, Warut and Kannan, Karthik Natarajan and Cohen, Maxime C., Addressing Fairness in Machine Learning Predictions: Strategic Best-Response Fair Discriminant Removed Algorithm (January 17, 2024). Available at SSRN: https://ssrn.com/abstract=3389631 or http://dx.doi.org/10.2139/ssrn.3389631

Hajime Shimao

Santa Fe Institute ( email )

1399 Hyde Park Road
Santa Fe, NM 87501
United States

Warut Khern-am-nuai (Contact Author)

McGill University - Desautels Faculty of Management ( email )

1001 Sherbrooke St. West
Montreal, Quebec H3A1G5 H3A 2M1
Canada

Karthik Natarajan Kannan

Purdue University ( email )

Krannert School of Management
West Lafayette, IN 47907
United States

Maxime C. Cohen

Desautels Faculty of Management, McGill University ( email )

1001 Sherbrooke St. W
Montreal, Quebec H3A 1G5
Canada

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
447
Abstract Views
2,203
Rank
123,718
PlumX Metrics