87
0

A Flexible Fairness Framework with Surrogate Loss Reweighting for Addressing Sociodemographic Disparities

Abstract

This paper presents a new algorithmic fairness framework called α\boldsymbol{\alpha}-β\boldsymbol{\beta} Fair Machine Learning (α\boldsymbol{\alpha}-β\boldsymbol{\beta} FML), designed to optimize fairness levels across sociodemographic attributes. Our framework employs a new family of surrogate loss functions, paired with loss reweighting techniques, allowing precise control over fairness-accuracy trade-offs through tunable hyperparameters α\boldsymbol{\alpha} and β\boldsymbol{\beta}. To efficiently solve the learning objective, we propose Parallel Stochastic Gradient Descent with Surrogate Loss (P-SGD-S) and establish convergence guarantees for both convex and nonconvex loss functions. Experimental results demonstrate that our framework improves overall accuracy while reducing fairness violations, offering a smooth trade-off between standard empirical risk minimization and strict minimax fairness. Results across multiple datasets confirm its adaptability, ensuring fairness improvements without excessive performance degradation.

View on arXiv
@article{xu2025_2503.16836,
  title={ A Flexible Fairness Framework with Surrogate Loss Reweighting for Addressing Sociodemographic Disparities },
  author={ Wen Xu and Elham Dolatabadi },
  journal={arXiv preprint arXiv:2503.16836},
  year={ 2025 }
}
Comments on this paper