Can I read Moss: Proxy Model-based Full-Weight Aggregation in Federated Learning with Heterogeneous Models on EtoBox?

Moss: Proxy Model-based Full-Weight Aggregation in Federated Learning with Heterogeneous Models by Cai, Yifeng; Zhang, Ziqi; Li, Ding; Guo, Yao; Chen, Xiangqun is a scholarly article available to read on EtoBox.

What is Moss: Proxy Model-based Full-Weight Aggregation in Federated Learning with Heterogeneous Models about?

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.

Author

Cai, Yifeng; Zhang, Ziqi; Li, Ding; Guo, Yao; Chen, Xiangqun

Published

2025

Language

EN