[2604.02017] Demographic Parity Tails for Regression

Statistics > Machine Learning arXiv:2604.02017 (stat) [Submitted on 2 Apr 2026] Title:Demographic Parity Tails for Regression Authors:Naht Sinh Le (LAMA), Christophe Denis (SAMM), Mohamed Hebiri (LAMA) View a PDF of the paper titled Demographic Parity Tails for Regression, by Naht Sinh Le (LAMA) and 2 other authors View PDF Abstract:Demographic parity (DP) is a widely studied fairness criterion in regression, enforcing independence between the predictions and sensitive attributes. However, constraining the entire distribution can degrade predictive accuracy and may be unnecessary for many applications, where fairness concerns are localized to specific regions of the distribution. To overcome this issue, we propose a new framework for regression under DP that focuses on the tails of target distribution across sensitive groups. Our methodology builds on optimal transport theory. By enforcing fairness constraints only over targeted regions of the distribution, our approach enables more nuanced and context-sensitive interventions. Leveraging recent advances, we develop an interpretable and flexible algorithm that leverages the geometric structure of optimal transport. We provide theoretical guarantees, including risk bounds and fairness properties, and validate the method through experiments in regression settings. Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG) Cite as: arXiv:2604.02017 [stat.ML]   (or arXiv:2604.02017v1 [stat.ML] for this version)   https://do...