A Convex Framework for Fair Regression

2017/06/07 by Richard A. Berk, Berk, Richard, Hoda Heidari +13 · 22 citations
Social Sciences · #Ethics and Social Impacts of AI #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · pdf · doi:10.48550/arxiv.1706.02409

openalex publication_date 2017/06/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We introduce a flexible family of fairness regularizers for (linear and logistic) regression problems. These regularizers all enjoy convexity, permitting fast optimization, and they span the rang from notions of group fairness to strong individual fairness. By varying the weight on the fairness regularizer, we can compute the efficient frontier of the accuracy-fairness trade-off on any given dataset, and we measure the severity of this trade-off via a numerical quantity we call the Price of Fairness (PoF). The centerpiece of our results is an extensive comparative study of the PoF across six different datasets in which fairness is a primary consideration.

Citations

• Equality of Opportunity in Supervised Learning
• Learning Non-Discriminatory Predictors
• Inherent Trade-Offs in the Fair Determination of Risk Scores

Cited by

• Penalized Fair Regression for Multiple Groups in Chronic Kidney Disease
• A Unifying Human-Centered AI Fairness Framework
• FairMT: Fairness for Heterogeneous Multi-Task Learning
• Measuring What Matters: Connecting AI Ethics Evaluations to System Attributes, Hazards, and Harms
• Wasserstein projection distance for fairness testing of regression models
• The Statistical Fairness-Accuracy Frontier
• Machine Learning with Multitype Protected Attributes: Intersectional Fairness through Regularisation
• A Comprehensive Evaluation of the Sensitivity of Density-Ratio Estimation Based Fairness Measurement in Regression
• Towards Accuracy-Fairness Paradox: Adversarial Example-based Data Augmentation for Visual Debiasing
• Pairwise Fairness for Ordinal Regression
• Transitioning from Real to Synthetic data: Quantifying the bias in model
• FairST: Equitable Spatial and Temporal Demand Prediction for New\n Mobility Systems
• Learning Fair and Interpretable Representations via Linear\n Orthogonalization
• Fair Algorithms for Learning in Allocation Problems
• Fair Forests: Regularized Tree Induction to Minimize Model Bias
• Calibration for the (Computationally-Identifiable) Masses
• Grading video interviews with fairness considerations
• Fairness and Missing Values
• Fair Decision Rules for Binary Classification
• A Primal-Dual Subgradient Approachfor Fair Meta Learning
• Monitoring of Static Fairness
• Modeling Techniques for Machine Learning Fairness: A Survey

Related

• Fair and Accurate Regression: Strong Formulations and Algorithms
• Fair Generalized Linear Models with a Convex Penalty
• Fair Regression: Quantitative Definitions and Reduction-based Algorithms
• fairret: a Framework for Differentiable Fairness Regularization Terms
• A minimax framework for quantifying risk-fairness trade-off in regression