Model-agnostic bias mitigation methods with regressor distribution control for Wasserstein-based fairness metrics

2021/11/19 by Alexey Miroshnikov, Miroshnikov, Alexey, Konstandinos Kotsiopoulos +5
Economics, Econometrics and Finance · Social Sciences · #49Q22 #68T01 #91A12 #Ethics and Social Impacts of AI #FOS: Computer and information sciences #FOS: Mathematics #Health Systems, Economic Evaluations, Quality of Life #Machine Learning (cs.LG) #Probability (math.PR)

paper · pdf · doi:10.48550/arxiv.2111.11259

openalex publication_date 2021/11/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

This article is a companion paper to our earlier work Miroshnikov et al. (2021) on fairness interpretability, which introduces bias explanations. In the current work, we propose a bias mitigation methodology based upon the construction of post-processed models with fairer regressor distributions for Wasserstein-based fairness metrics. By identifying the list of predictors contributing the most to the bias, we reduce the dimensionality of the problem by mitigating the bias originating from those predictors. The post-processing methodology involves reshaping the predictor distributions by balancing the positive and negative bias explanations and allows for the regressor bias to decrease. We design an algorithm that uses Bayesian optimization to construct the bias-performance efficient frontier over the family of post-processed models, from which an optimal model is selected. Our novel methodology performs optimization in low-dimensional spaces and avoids expensive model retraining.

Related

• Explainable post-training bias mitigation with distribution-based fairness metrics
• Post-processing fairness with minimal changes
• Fairness by Explicability and Adversarial SHAP Learning
• Bias Mitigation in Fine-tuning Pre-trained Models for Enhanced Fairness and Efficiency
• Wasserstein projection distance for fairness testing of regression models