Federated Learning Over Wireless Channels: Dynamic Resource Allocation and Task Scheduling

TL;DR

This paper develops online stochastic learning algorithms to optimize resource allocation and task scheduling in federated learning over wireless channels, improving training efficiency under energy and channel constraints.

Contribution

It introduces a novel CMDP model for FL over wireless channels and proposes online algorithms with proven convergence to enhance performance.

Findings

Achieves better training performance than benchmark algorithms.

Effectively manages long-term energy and channel constraints.

Provides convergence proof for the proposed algorithms.

Abstract

With the development of federated learning (FL), mobile devices (MDs) are able to train their local models with private data and sends them to a central server for aggregation, thereby preventing sensitive raw data leakage. In this paper, we aim to improve the training performance of FL systems in the context of wireless channels and stochastic energy arrivals of MDs. To this purpose, we dynamically optimize MDs' transmission power and training task scheduling. We first model this dynamic programming problem as a constrained Markov decision process (CMDP). Due to high dimensions rooted from our CMDP problem, we propose online stochastic learning methods to simplify the CMDP and design online algorithms to obtain an efficient policy for all MDs. Since there are long-term constraints in our CMDP, we utilize Lagrange multipliers approach to tackle this issue. Furthermore, we prove the convergence of the proposed online stochastic learning algorithm. Numerical results indicate that the proposed algorithms can achieve better performance than the benchmark algorithms.