Privacy-Aware Federated Learning with Adaptive User Selection

Federated learning (FL) enables multiple edge devices to collaboratively train a machine learning model without the need to share potentially private data. Federated learning proceeds through iterative exchanges of model updates, which pose two key challenges: (i) the accumulation of privacy leakage over time and (ii) communication latency. These two limitations are typically addressed separately: (i) via perturbed updates to enhance privacy and (ii) user selection to mitigate latency both at the expense of accuracy. In this work, we propose a method that jointly addresses the accumulation of privacy leakage and communication latency via active user selection, aiming to improve the trade-off among privacy, latency, and model performance. To achieve this, we construct a reward function that accounts for these three objectives. Building on this reward, we propose a multi-armed bandit (MAB)-based algorithm, termed privacy-aware active user selection (PAUSE) – which dynamically selects a subset of users each round while ensuring bounded overall privacy leakage. We establish a theoretical analysis, systematically showing that the regret growth rate of PAUSE follows that of the bestknown rate in MAB literature. To address the complexity overhead of active user selection, we propose a simulated annealing-based relaxation of PAUSE and analyze its ability to approximate the rewardmaximizing policy under reduced complexity. We numerically validate the privacy leakage, associated improved latency, and accuracy gains of our methods for the federated training in various scenarios.