Communication-Efficient Device Scheduling for Federated Learning Using Lyapunov Optimization

2025-03-01

IEEE Transactions on Networking

AI Summary

To address slow convergence in federated learning (FL) under wireless resource constraints—caused by intermittent device connectivity, heterogeneous channel conditions, and non-i.i.d. data—this paper proposes a joint online client selection and power allocation framework. Methodologically, it innovatively integrates Lyapunov drift-plus-penalty theory into FL scheduling, requiring only instantaneous channel state information (CSI) without prior knowledge of channel statistics. It further introduces a participation-probability-weighted gradient aggregation scheme, for which we establish theoretical convergence to stationary points under non-convex, non-i.i.d. settings—unifying both full and optimal partial participation regimes. Experimental evaluation on heterogeneous CIFAR-10 data demonstrates that the proposed method significantly reduces average communication time compared to random participation, with particularly pronounced gains under highly disparate channel qualities. These results validate its efficiency and robustness under realistic wireless constraints.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Non-convex Optimization

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Federated learning (FL) is a useful tool that enables the training of machine learning models over distributed data without having to collect data centrally. When deploying FL in constrained wireless environments, however, intermittent connectivity of devices, heterogeneous connection quality, and non-i.i.d. data can severely slow convergence. In this paper, we consider FL with arbitrary device participation probabilities for each round and show that by weighing each device's update by the reciprocal of their per-round participation probability, we can guarantee convergence to a stationary point. Our bound applies to non-convex loss functions and non-i.i.d. datasets and recovers state-of-the-art convergence rates for both full and uniform partial participation, including linear speedup, with only a single-sided learning rate. Then, using the derived convergence bound, we develop a new online client selection and power allocation algorithm that utilizes the Lyapunov drift-plus-penalty framework to opportunistically minimize a function of the convergence bound and the average communication time under a transmit power constraint. We use optimization over manifold techniques to obtain a solution to the minimization problem. Thanks to the Lyapunov framework, one key feature of the algorithm is that knowledge of the channel distribution is not required and only the instantaneous channel state information needs to be known. Using the CIFAR-10 dataset with varying levels of data heterogeneity, we show through simulations that the communication time can be significantly decreased using our algorithm compared to uniformly random participation, especially for heterogeneous channel conditions.

Problem

• Optimizes device scheduling in federated learning for faster convergence.
• Addresses non-i.i.d. data and intermittent connectivity in wireless environments.
• Minimizes communication time and power usage using Lyapunov optimization.

Innovation

• Lyapunov drift-plus-penalty for client selection
• Optimization over manifold for power allocation
• Reciprocal weighting for device participation probability