FairReweighing: Density Estimation-Based Reweighing Framework for Improving Separation in Fair Regression
Authors:
- Xiaoyin XiXXXiaoyin Xi
- Zhe YuZYZhe Yu
Preprint
Abstract
There has been a prevalence of applying AI software in both high-stakes public-sector and industrial contexts. However, the lack of transparency has raised concerns about whether these data-informed AI software decisions secure fairness against people of all racial, gender, or age groups. Despite extensive research on emerging fairness-aware AI software, up to now most efforts to solve this issue have been dedicated to binary classification tasks. Fairness in regression is relatively underexplored. In this work, we adopted a mutual information-based metric to assess separation violations. The metric is also extended so that it can be directly applied to both classification and regression problems with both binary and continuous sensitive attributes. Inspired by the Reweighing algorithm in fair classification, we proposed a FairReweighing pre-processing algorithm based on density estimation to ensure that the learned model satisfies the separation criterion. Theoretically, we show that the proposed FairReweighing algorithm can guarantee separation in the training data under a data independence assumption. Empirically, on both synthetic and real-world data, we show that FairReweighing outperforms existing state-of-the-art regression fairness solutions in terms of improving separation while maintaining high accuracy.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- FairBalance: How to Achieve Equalized Odds With Data Pre-ProcessingIEEE Transactions on Software Engineering, 2024
- FairMask: Better Fairness via Model-Based Rebalancing of Protected AttributesIEEE Transactions on Software Engineering, 2022
- Bias in machine learning software: why? how? what to do?2021
- A Survey on Bias and Fairness in Machine LearningACM Computing Surveys, 2021
- A snapshot of the frontiers of fairness in machine learningCommunications of the ACM, 2020
- Semantics derived automatically from language corpora contain human-like biasesScience, 2017
- Machine learning for targeted display advertising: transfer learning in actionMachine Learning, 2013
- Data preprocessing techniques for classification without discriminationKnowledge and Information Systems, 2011
- Consumer credit-risk models via machine-learning algorithmsJournal of Banking & Finance, 2010
- A data-driven software tool for enabling cooperative information sharing among police departmentsEuropean Journal of Operational Research, 2002