An Efficient Asynchronous Federated Learning Protocol for Edge Devices

Recent studies highlight the significant potential of edge computing and federated learning (FL) in advancing artificial intelligence. However, challenges, such as unstable device performance and the heterogeneously distributed feature of local data, pose threats to the efficiency of global model training. To address these issues, we propose an efficient federated edge learning protocol with key innovations: 1) propose a timing query mechanism, which controls the impact of slow clients on interaction time and changes the parameter server from a passive receiver to an active querier to guarantee the aggregated subset size; 2) propose a screening supplementary strategy from the perspective of optimizing the quality of the data combination; and 3) integrate the timing query mechanism and screening supplementary strategy in a flexible manner. This combination improves the interaction efficiency of global model training, focusing on both interaction time and aggregation update quality. Experiments show that while ensuring the convergence efficiency of each round and the global model performance, the interaction efficiency has been improved by 25% to 75%, thus ensuring the efficiency of FL.

I. Introduction

Federated learning (FL) is composed of a parameter server and multiple edge devices, achieving the updating and optimization of a global model through the aggregation calculation of model parameters [1]. As shown in Fig. 1, FL enables cross-device model training while ensuring privacy protection by keeping the original data on the local devices. However, local clients are progressively expanding to various small-edge devices beyond servers and base stations [2], [3], such as smartphones and wearable devices [4], [5]. This expansion not only results in an increased number of clients but also introduces a rise in diversity concerning computing power, storage, and electricity, so heterogeneity poses a more prominent bottleneck than communication in the current generation of battery-powered mobile devices [6]. Fig. 1.

Training process of FL.