Advanced Optimization Techniques for Federated Learning on Non-IID Data
Future Internet · 13.10.2024 · 16 (10) · 370
DOI: 10.3390/fi16100370 ↗Kurz gefasst
Fachartikel in der Zeitschrift Future Internet, online erschienen am 13.10.2024. Verfasst von Filippos Efthymiadis und 3 weiteren Autor:innen. Die Arbeit wurde bisher 23-mal zitiert (Crossref) und nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Federated learning enables model training on multiple clients locally, without the need to transfer their data to a central server, thus ensuring data privacy. In this paper, we investigate the impact of Non-Independent and Identically Distributed (non-IID) data on the performance of federated training, where we find a reduction in accuracy of up to 29% for neural networks trained in environments with skewed non-IID data. Two optimization strategies are presented to address this issue. The first strategy focuses on applying a cyclical learning rate to determine the learning rate during federated training, while the second strategy develops a sharing and pre-training method on augmented data in order to improve the efficiency of the algorithm in the case of non-IID data. By combining these two methods, experiments show that the accuracy on the CIFAR-10 dataset increased by about 36% while achieving faster convergence by reducing the number of required communication rounds by 5.33 times. The proposed techniques lead to improved accuracy and faster model convergence, thus representing a significant advance in the field of federated learning and facilitating its application to real-world scenarios.
Zitieren
APA
Efthymiadis, F., Karras, A., Karras, C., & Sioutas, S. (2024). Advanced Optimization Techniques for Federated Learning on Non-IID Data. Future Internet, 16(10), 370. https://doi.org/10.3390/fi16100370
Vancouver
Efthymiadis F, Karras A, Karras C, Sioutas S. Advanced Optimization Techniques for Federated Learning on Non-IID Data. Future Internet. 2024;16(10):370. doi:10.3390/fi16100370
BibTeX
@article{Efthymiadis2024,
author = {Efthymiadis, Filippos and Karras, Aristeidis and Karras, Christos and Sioutas, Spyros},
title = {Advanced Optimization Techniques for Federated Learning on Non-IID Data},
journal = {Future Internet},
year = {2024},
volume = {16},
number = {10},
pages = {370},
doi = {10.3390/fi16100370}
}
EUVIMED