Utilize este identificador para referenciar este registo:
https://hdl.handle.net/10316/114426
Título: | FAIR-FATE: Fair Federated Learning with Momentum | Autor: | Salazar, Teresa Fernandes, Miguel Araújo, Helder Abreu, Pedro Henriques |
Palavras-chave: | Fairness; Federated Learning; Machine Learning; Momentum | Data: | 2023 | Editora: | Springer Nature | Projeto: | UIDB/00326/2020 UIDP/00326/2020 Research Grants 2021.05763.BD |
Título da revista, periódico, livro ou evento: | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Volume: | 14073 | Resumo: | While fairness-aware machine learning algorithms have been receiving increasing attention, the focus has been on centralized machine learning, leaving decentralized methods underexplored. Federated Learning is a decentralized form of machine learning where clients train local models with a server aggregating them to obtain a shared global model. Data heterogeneity amongst clients is a common characteristic of Federated Learning, which may induce or exacerbate discrimination of unprivileged groups defined by sensitive attributes such as race or gender. In this work we propose FAIR-FATE: a novel FAIR FederATEd Learning algorithm that aims to achieve group fairness while maintaining high utility via a fairness-aware aggregation method that computes the global model by taking into account the fairness of the clients. To achieve that, the global model update is computed by estimating a fair model update using a Momentum term that helps to overcome the oscillations of nonfair gradients. To the best of our knowledge, this is the first approach in machine learning that aims to achieve fairness using a fair Momentum estimate. Experimental results on real-world datasets demonstrate that FAIR-FATE outperforms state-of-the-art fair Federated Learning algorithms under different levels of data heterogeneity | URI: | https://hdl.handle.net/10316/114426 | ISSN: | 0302-9743 1611-3349 |
DOI: | 10.1007/978-3-031-35995-8_37 | Direitos: | openAccess |
Aparece nas coleções: | FCTUC Eng.Electrotécnica - Artigos em Revistas Internacionais I&D ISR - Artigos em Revistas Internacionais FCTUC Eng.Informática - Artigos em Revistas Internacionais I&D CISUC - Artigos em Revistas Internacionais |
Ficheiros deste registo:
Ficheiro | Descrição | Tamanho | Formato | |
---|---|---|---|---|
preprint_FAIR-FATE Fair Federated Learning with Momentum.pdf | 1.07 MB | Adobe PDF | Ver/Abrir |
Visualizações de página
83
Visto em 2/out/2024
Downloads
36
Visto em 2/out/2024
Google ScholarTM
Verificar
Altmetric
Altmetric
Todos os registos no repositório estão protegidos por leis de copyright, com todos os direitos reservados.