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

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Shi, Enze, Kong, Linglong, Jiang, Bei
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916680155791360
author Shi, Enze
Kong, Linglong
Jiang, Bei
author_facet Shi, Enze
Kong, Linglong
Jiang, Bei
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2504_06470
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Deep Fair Learning: A Unified Framework for Fine-tuning Representations with Sufficient Networks
Shi, Enze
Kong, Linglong
Jiang, Bei
Machine Learning
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.
title Deep Fair Learning: A Unified Framework for Fine-tuning Representations with Sufficient Networks
topic Machine Learning
url https://arxiv.org/abs/2504.06470