FedNolowe: A normalized loss-based weighted aggregation strategy for robust federated learning in heterogeneous environments
PLOS One · 14.08.2025 · 20 (8) · e0322766
DOI: 10.1371/journal.pone.0322766 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 14.08.2025. Verfasst von Duy-Dong Le und 4 weiteren Autor:innen. Beteiligte Einrichtung: Industrial University of Ho Chi Minh City (IUH), Ho Chi Minh City, Vietnam. Hauptthemen laut MEDLINE: Software, Machine Learning. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Federated Learning supports collaborative model training across distributed clients while keeping sensitive data decentralized. Still, non-independent and identically distributed data pose challenges like unstable convergence and client drift. We propose Federated Normalized Loss-based Weighted Aggregation (FedNolowe) (Code is available at https://github.com/dongld-2020/fednolowe ), a new method that weights client contributions using normalized training losses, favoring those with lower losses to improve global model stability. Unlike prior methods tied to dataset sizes or resource-heavy techniques, FedNolowe employs a two-stage L1 normalization, reducing computational complexity by 40% in floating-point operations while matching state-of-the-art performance. A detailed sensitivity analysis shows our two-stage weighting maintains stability in heterogeneous settings by mitigating extreme loss impacts while remaining effective in independent and identically distributed scenarios.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Industrial University of Ho Chi Minh City (IUH), Ho Chi Minh City, Vietnam.
- National Institute of Information and Communications Technology (NICT), Koganei, Japan.
- Saigon University (SGU), Ho Chi Minh City, Vietnam.
ORCID-Profile: Duy-Dong Le
Zitieren
APA
Le, D. D., Huynh, T. N., Tran, A. K., Dao, M. S., & Bao, P. T. (2025). FedNolowe: A normalized loss-based weighted aggregation strategy for robust federated learning in heterogeneous environments. PLOS One, 20(8), e0322766. https://doi.org/10.1371/journal.pone.0322766
Vancouver
Le DD, Huynh TN, Tran AK, Dao MS, Bao PT. FedNolowe: A normalized loss-based weighted aggregation strategy for robust federated learning in heterogeneous environments. PLOS One. 2025;20(8):e0322766. doi:10.1371/journal.pone.0322766
BibTeX
@article{Le2025,
author = {Le, Duy-Dong and Huynh, Tuong-Nguyen and Tran, Anh-Khoa and Dao, Minh-Son and Bao, Pham The},
title = {FedNolowe: A normalized loss-based weighted aggregation strategy for robust federated learning in heterogeneous environments},
journal = {PLOS One},
year = {2025},
volume = {20},
number = {8},
pages = {e0322766},
doi = {10.1371/journal.pone.0322766}
}
EUVIMED