Communication-efficient online federated composite optimization

doi:10.1016/j.automatica.2025.112679

Communication-efficient online federated composite optimization

2025-11-12

0

PRE

AI

DOI：10.1016/j.automatica.2025.112679

Original

Original request for help

Save

Abstract

En 中文

Online federated optimization is crucial for sequential decision-making in dynamic environments, yet it often overlooks non-smooth regularizers, which are common in real-world applications such as machine learning and wireless communication. Additionally, communication overhead is an important facet in practice. As such, we focus on addressing the two issues by introducing the online federated composite optimization problem, where the loss function is time-varying and contains a non-smooth regularizer, and employing compressors to reduce communication overhead. An algorithm, named FedOEC is proposed, which simultaneously resolves online optimization problems with non-smooth regularizers and reduces communication overhead by leveraging multi-kernel and compressors efficiently. Through theoretical analysis, FedOEC achieves optimal sublinear regret bound O(T) with time-varying step sizes in convex settings, where T represents the number of communication rounds. Finally, numerical experiments confirm the effectiveness of the proposed algorithm.

Keywords:

Federated composite optimization

Online learning

Compression

Multi-kernel learning

Regret

Communication-efficient online federated composite optimization

Abstract

Journal

Organization