Age of Information Based Client Selection for Wireless Federated Learning With Diversified Learning Capabilities

Federated Learning (FL) empowers wireless intelligent applications, by leveraging distributed data of edge clients for training without compromising privacy. Client selection is inevitable in FL, since clients have diversified learning capabilities arising from heterogeneous computing and communication resources. Existing methods like fair-selection and dropping-straggler are either inefficient or unfair (resulting in a less effective trained model). Therefore, we propose FedAoI, an Age-of-Information (AoI) based client selection policy. FedAoI ensures fairness by allowing all clients, including stragglers, to submit their model updates while maintaining high training efficiency by keeping round completion times short. This trade-off is achieved by minimizing Peak-AoI (PAoI), the interval between a client's consecutive participations. An optimization problem is formulated by minimizing the Expected-Weighted-Sum-of-PAoI. This NP-hard problem is addressed with a two-step sub-optimal algorithm, PriorS. It first calculates client priority in a round using Lyapunov optimization and then selects the highest-priority clients through G-FPFC (Greedy minimization of the round weighted-sum-of-PAoI with First-Priority-First-Considered). Simulation results demonstrate that, compared to fair-selection, FedAoI improves average efficiency by 83.8% and achieves an average model accuracy of 97.3% (or at the cost of averaging 2.7% degradation in model accuracy). Compared to dropping-straggler, FedAoI reduces the average model accuracy degradation from 9.5% to 2.7%.

I. Introduction

With the rapid development of wireless communications [1], [2], [3], [4], [5], the era of Internet of Things is coming with the sixth generation of mobile networks (6G) [6], [7]. Artificial intelligence (AI) is a key technology for 6G [8], [9], enabling a wide variety of wireless intelligent applications spanning from speech recognition, computer vision to automation [10]. Big data is one of the key drivers that boost AI performance. With the rapid development of processor chips, mobile clients with integrated capabilities of sensing, communication, and computing can generate useful data, which is emerging as a major source to promote the performance of AI applications [11]. Traditionally, the raw data stored in a wide range of clients is collected by companies or institutions. However, this centralized framework for gathering and training has an Achilles’ heel, i.e., it violates data privacy by exposing raw data to the gatherers. Moreover, the available data for centralized training is limited as clients may be reluctant to upload data, constraining model performance. To solve this problem, Google proposed Federated Learning (FL), which is a distributed learning framework over plenty of clients connected to edge networks [12]. The key idea of FL is to train a Machine Learning (ML) model using local computing resources and data without transferring data elsewhere. FL is qualified to guarantee data privacy, while abundant data from extensive clients could be leveraged to improve the model performance. FL is promising to advance intelligent applications and has numerous successful explorations and implementations in IoT [13], [14], [15], intelligent core networks [16], self-driving vehicles [17], [18], and smart health [19], [20], [21], etc.