Deep Fair Learning: A Unified Framework for Fine-tuning Representations with Sufficient Networks

2025/04/08 by Enze Shi, Shi, Enze, Linglong Kong +3 · 1 citation
Computer Science · Social Sciences · #Adversarial Robustness in Machine Learning #Ethics and Social Impacts of AI #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · pdf · doi:10.48550/arxiv.2504.06470

openalex publication_date 2025/04/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Ensuring fairness in machine learning is a critical and challenging task, as biased data representations often lead to unfair predictions. To address this, we propose Deep Fair Learning, a framework that integrates nonlinear sufficient dimension reduction with deep learning to construct fair and informative representations. By introducing a novel penalty term during fine-tuning, our method enforces conditional independence between sensitive attributes and learned representations, addressing bias at its source while preserving predictive performance. Unlike prior methods, it supports diverse sensitive attributes, including continuous, discrete, binary, or multi-group types. Experiments on various types of data structure show that our approach achieves a superior balance between fairness and utility, significantly outperforming state-of-the-art baselines.

Cited by

• Understanding Fairness and Prediction Error through Subspace Decomposition and Influence Analysis

Related

• Last-Layer Fairness Fine-tuning is Simple and Effective for Neural Networks
• Learning Fairer Representations with FairVIC
• Towards Auditability for Fairness in Deep Learning
• 10 Years of Fair Representations: Challenges and Opportunities
• FairNN- Conjoint Learning of Fair Representations for Fair Decisions