Achieving Equity via Transfer Learning with Fairness Optimization

Xiaoyang Wang; Chia-Hsuan Chang; Christopher C. Yang

doi:10.1109/ACCESS.2024.3519465

Back

Achieving Equity via Transfer Learning with Fairness Optimization

Journal article

Open access

Peer reviewed

Achieving Equity via Transfer Learning with Fairness Optimization

Xiaoyang Wang, Chia-Hsuan Chang and Christopher C. Yang

IEEE access, v 12

01 Jan 2024

DOI: https://doi.org/10.1109/ACCESS.2024.3519465

Files and links (1)

url

https://doi.org/10.1109/ACCESS.2024.3519465View

Published, Version of Record (VoR) Open

Abstract

Accuracy

Adaptation models

AI Fairness

Data models

In-processing Bias Mitigation

Machine learning algorithms

Performance–Fairness

Predictive models

Prevention and mitigation

Trade-off

Training

Transfer learning

Transforms

Optimization

Machine learning algorithms are increasingly used in real-world decision-making systems, raising concerns about potential biases and unfairness. Existing in-processing bias mitigation approaches often focus on achieving numerical parity across demographic groups while neglecting the performance impact on individual groups and the overall performance-fairness trade-off. This can lead to increased discriminatory outcomes and hinder efforts to develop truly fair AI systems. This paper proposes Transfer Learning with Fairness Optimization (TLFO), a novel framework that serially optimizes predictive performance and fairness in machine learning models. TLFO leverages transfer learning by dividing the training process into two distinct phases: (1) initial learning for performance optimization and (2) subsequent finetuning for fairness enhancement. This sequential approach enables fine-grained control over fairness constraints, minimizing the performance-fairness trade-off. Extensive experiments on two real-world datasets demonstrate TLFO's effectiveness. TLFO consistently achieves superior fairness with minimal performance degradation compared to state-of-the-art in-processing bias mitigation approaches, highlighting its potential for generating fair and accurate classifiers with versatile applications.

Metrics

4 Record Views

Details

Title: Achieving Equity via Transfer Learning with Fairness Optimization
Creators: Xiaoyang Wang - Drexel University
Chia-Hsuan Chang - College of Computing & Informatics, Drexel University, Philadelphia, PA, USA
Christopher C. Yang - Drexel University
Publication Details: IEEE access, v 12
Publisher: IEEE
Grant note: IIS-1741306; IIS-2235548 / National Science Foundation (10.13039/100000001) DoD W91XWH-05-1-023 / U.S. Department of Defense (10.13039/100000005)
Resource Type: Journal article
Language: English
Academic Unit: Information Science (Informatics); College of Computing and Informatics
Scopus ID: 2-s2.0-85212773979
Other Identifier: 991022008196604721

Achieving Equity via Transfer Learning with Fairness Optimization

Files and links (1)

Abstract

Metrics

Details

Drexel University Social media