Age-Based Scheduling Policy for Federated Learning in Mobile Edge\n Networks

Federated learning (FL) is a machine learning model that preserves data\nprivacy in the training process. Specifically, FL brings the model directly to\nthe user equipments (UEs) for local training, where an edge server periodically\ncollects the trained parameters to produce an improved model and sends it back\nto the UEs. However, since communication usually occurs through a limited\nspectrum, only a portion of the UEs can update their parameters upon each\nglobal aggregation. As such, new scheduling algorithms have to be engineered to\nfacilitate the full implementation of FL. In this paper, based on a metric\ntermed the age of update (AoU), we propose a scheduling policy by jointly\naccounting for the staleness of the received parameters and the instantaneous\nchannel qualities to improve the running efficiency of FL. The proposed\nalgorithm has low complexity and its effectiveness is demonstrated by Monte\nCarol simulations.\n