Moss: Proxy Model-based Full-Weight Aggregation in Federated Learning with Heterogeneous Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cai, Yifeng, Zhang, Ziqi, Li, Ding, Guo, Yao, Chen, Xiangqun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917955625811968
author Cai, Yifeng
Zhang, Ziqi
Li, Ding
Guo, Yao
Chen, Xiangqun
author_facet Cai, Yifeng
Zhang, Ziqi
Li, Ding
Guo, Yao
Chen, Xiangqun
contents Modern Federated Learning (FL) has become increasingly essential for handling highly heterogeneous mobile devices. Current approaches adopt a partial model aggregation paradigm that leads to sub-optimal model accuracy and higher training overhead. In this paper, we challenge the prevailing notion of partial-model aggregation and propose a novel "full-weight aggregation" method named Moss, which aggregates all weights within heterogeneous models to preserve comprehensive knowledge. Evaluation across various applications demonstrates that Moss significantly accelerates training, reduces on-device training time and energy consumption, enhances accuracy, and minimizes network bandwidth utilization when compared to state-of-the-art baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2503_10218
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Moss: Proxy Model-based Full-Weight Aggregation in Federated Learning with Heterogeneous Models
Cai, Yifeng
Zhang, Ziqi
Li, Ding
Guo, Yao
Chen, Xiangqun
Machine Learning
Cryptography and Security
Modern Federated Learning (FL) has become increasingly essential for handling highly heterogeneous mobile devices. Current approaches adopt a partial model aggregation paradigm that leads to sub-optimal model accuracy and higher training overhead. In this paper, we challenge the prevailing notion of partial-model aggregation and propose a novel "full-weight aggregation" method named Moss, which aggregates all weights within heterogeneous models to preserve comprehensive knowledge. Evaluation across various applications demonstrates that Moss significantly accelerates training, reduces on-device training time and energy consumption, enhances accuracy, and minimizes network bandwidth utilization when compared to state-of-the-art baselines.
title Moss: Proxy Model-based Full-Weight Aggregation in Federated Learning with Heterogeneous Models
topic Machine Learning
Cryptography and Security
url https://arxiv.org/abs/2503.10218