FedGiA: An Efficient Hybrid Algorithm for Federated Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2022
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911845307121664 |
|---|---|
| author | Zhou, Shenglong Li, Geoffrey Ye |
| author_facet | Zhou, Shenglong Li, Geoffrey Ye |
| contents | Federated learning has shown its advances recently but is still facing many challenges, such as how algorithms save communication resources and reduce computational costs, and whether they converge. To address these critical issues, we propose a hybrid federated learning algorithm (FedGiA) that combines the gradient descent and the inexact alternating direction method of multipliers. The proposed algorithm is more communication- and computation-efficient than several state-of-the-art algorithms theoretically and numerically. Moreover, it also converges globally under mild conditions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2205_01438 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | FedGiA: An Efficient Hybrid Algorithm for Federated Learning Zhou, Shenglong Li, Geoffrey Ye Machine Learning Optimization and Control Federated learning has shown its advances recently but is still facing many challenges, such as how algorithms save communication resources and reduce computational costs, and whether they converge. To address these critical issues, we propose a hybrid federated learning algorithm (FedGiA) that combines the gradient descent and the inexact alternating direction method of multipliers. The proposed algorithm is more communication- and computation-efficient than several state-of-the-art algorithms theoretically and numerically. Moreover, it also converges globally under mild conditions. |
| title | FedGiA: An Efficient Hybrid Algorithm for Federated Learning |
| topic | Machine Learning Optimization and Control |
| url | https://arxiv.org/abs/2205.01438 |