Achieving Equity via Transfer Learning With Fairness 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 impa...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10804762/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841533432087183360 |
|---|---|
| author | Xiaoyang Wang Chia-Hsuan Chang Christopher C. Yang |
| author_facet | Xiaoyang Wang Chia-Hsuan Chang Christopher C. Yang |
| author_sort | Xiaoyang Wang |
| collection | DOAJ |
| description | 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: <xref ref-type="disp-formula" rid="deqn1">(1)</xref> initial learning for performance optimization and <xref ref-type="disp-formula" rid="deqn2">(2)</xref> subsequent fine-tuning 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. |
| format | Article |
| id | doaj-art-af3da2765d0b49e1b6d1bb9bcebb8aa4 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-af3da2765d0b49e1b6d1bb9bcebb8aa42025-01-16T00:01:33ZengIEEEIEEE Access2169-35362024-01-011219522919524110.1109/ACCESS.2024.351946510804762Achieving Equity via Transfer Learning With Fairness OptimizationXiaoyang Wang0https://orcid.org/0000-0002-8471-4670Chia-Hsuan Chang1https://orcid.org/0000-0001-9116-8244Christopher C. Yang2https://orcid.org/0000-0001-5463-6926College of Computing & Informatics, Drexel University, Philadelphia, PA, USACollege of Computing & Informatics, Drexel University, Philadelphia, PA, USACollege of Computing & Informatics, Drexel University, Philadelphia, PA, USAMachine 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: <xref ref-type="disp-formula" rid="deqn1">(1)</xref> initial learning for performance optimization and <xref ref-type="disp-formula" rid="deqn2">(2)</xref> subsequent fine-tuning 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.https://ieeexplore.ieee.org/document/10804762/AI fairnesstransfer learningin-processing bias mitigationperformance-fairness trade-off |
| spellingShingle | Xiaoyang Wang Chia-Hsuan Chang Christopher C. Yang Achieving Equity via Transfer Learning With Fairness Optimization IEEE Access AI fairness transfer learning in-processing bias mitigation performance-fairness trade-off |
| title | Achieving Equity via Transfer Learning With Fairness Optimization |
| title_full | Achieving Equity via Transfer Learning With Fairness Optimization |
| title_fullStr | Achieving Equity via Transfer Learning With Fairness Optimization |
| title_full_unstemmed | Achieving Equity via Transfer Learning With Fairness Optimization |
| title_short | Achieving Equity via Transfer Learning With Fairness Optimization |
| title_sort | achieving equity via transfer learning with fairness optimization |
| topic | AI fairness transfer learning in-processing bias mitigation performance-fairness trade-off |
| url | https://ieeexplore.ieee.org/document/10804762/ |
| work_keys_str_mv | AT xiaoyangwang achievingequityviatransferlearningwithfairnessoptimization AT chiahsuanchang achievingequityviatransferlearningwithfairnessoptimization AT christophercyang achievingequityviatransferlearningwithfairnessoptimization |