Search papers, labs, and topics across Lattice.
This paper addresses the challenge of privacy heterogeneity in differentially private federated learning (DP-FL) where clients have varying privacy budgets. They derive a convergence analysis that explicitly quantifies the impact of privacy heterogeneity on the training error in DP-FL. Based on this analysis, they propose a privacy-aware client selection strategy formulated as a convex optimization problem to minimize training error by adaptively adjusting client selection probabilities.
Ignoring privacy differences between clients in federated learning can hurt accuracy by 10%, but a new privacy-aware client selection method fixes this.
Differentially private federated learning (DP-FL) enables clients to collaboratively train machine learning models while preserving the privacy of their local data. However, most existing DP-FL approaches assume that all clients share a uniform privacy budget, an assumption that does not hold in real-world scenarios where privacy requirements vary widely. This privacy heterogeneity poses a significant challenge: conventional client selection strategies, which typically rely on data quantity, cannot distinguish between clients providing high-quality updates and those introducing substantial noise due to strict privacy constraints. To address this gap, we present the first systematic study of privacy-aware client selection in DP-FL. We establish a theoretical foundation by deriving a convergence analysis that quantifies the impact of privacy heterogeneity on training error. Building on this analysis, we propose a privacy-aware client selection strategy, formulated as a convex optimization problem, that adaptively adjusts selection probabilities to minimize training error. Extensive experiments on benchmark datasets demonstrate that our approach achieves up to a 10% improvement in test accuracy on CIFAR-10 compared to existing baselines under heterogeneous privacy budgets. These results highlight the importance of incorporating privacy heterogeneity into client selection for practical and effective federated learning.
