2025-01-14-xie25c.md

title	Hierarchical Global Asynchronous Federated Learning Across Multi-Center
booktitle	Proceedings of the 16th Asian Conference on Machine Learning
year	2025
volume	260
series	Proceedings of Machine Learning Research
month	0
publisher	PMLR
pdf	https://raw.githubusercontent.com/mlresearch/v260/main/assets/xie25c/xie25c.pdf
url	https://proceedings.mlr.press/v260/xie25c.html
openreview	MpXwb1SRWF
abstract	Federated learning for training machine learning models across geographically distributed regional centers is becoming prevalent. However, because of disparities in location, latency, and computational capabilities, synchronously aggregating models across different sites requires waiting for stragglers, leading to significant delays. Traditional asynchronous aggregation across regional centers still faces issues of stale model parameters and outdated gradients due to the hierarchical aggregation involving local clients within each center. To address this, we propose Hierarchical Global Asynchronous Federated Learning (HGA-FL), which combines global asynchronous model aggregation across regional centers with synchronous aggregation and local consistent regularization alignment within each local center. We theoretically analyze the convergence rate of our method under non-convex optimization settings, demonstrating its stable convergence during the aggregation. Experimental evaluations show that our approach outperforms other baseline two-level aggregation methods in terms of global model generalization ability, particularly under conditions of data heterogeneity, latency, and gradient staleness.
layout	inproceedings
issn	2640-3498
id	xie25c
tex_title	Hierarchical Global Asynchronous Federated Learning Across Multi-Center
firstpage	543
lastpage	558
page	543-558
order	543
cycles	false
bibtex_editor	Nguyen, Vu and Lin, Hsuan-Tien
editor	given family Vu Nguyen given family Hsuan-Tien Lin
bibtex_author	Xie, Wei and Xiong, Runqun and Luo, Junzhou
author	given family Wei Xie given family Runqun Xiong given family Junzhou Luo
date	2025-01-14
address
container-title	Proceedings of the 16th Asian Conference on Machine Learning
genre	inproceedings
issued	date-parts 2025 1 14
extras	label link Supplementary PDF https://raw.githubusercontent.com/mlresearch/v260/main/assets/assets/xie25c/xie25c-supp.pdf