Differentially Private Federated Learning With Time-Adaptive Privacy Spending

TL;DR

This paper introduces a time-adaptive differential privacy framework for federated learning, allowing clients to allocate privacy budgets unevenly over time, leading to improved utility and privacy trade-offs.

Contribution

It proposes a novel time-adaptive privacy spending strategy in federated learning, optimizing privacy budget allocation across rounds for better utility.

Findings

Theoretically proves utility improvements with uneven privacy budget spending.

Empirically demonstrates enhanced privacy-utility trade-offs on benchmark datasets.

Abstract

Federated learning (FL) with differential privacy (DP) provides a framework for collaborative machine learning, enabling clients to train a shared model while adhering to strict privacy constraints. The framework allows each client to have an individual privacy guarantee, e.g., by adding different amounts of noise to each client's model updates. One underlying assumption is that all clients spend their privacy budgets uniformly over time (learning rounds). However, it has been shown in the literature that learning in early rounds typically focuses on more coarse-grained features that can be learned at lower signal-to-noise ratios while later rounds learn fine-grained features that benefit from higher signal-to-noise ratios. Building on this intuition, we propose a time-adaptive DP-FL framework that expends the privacy budget non-uniformly across both time and clients. Our framework enables each client to save privacy budget in early rounds so as to be able to spend more in later rounds when additional accuracy is beneficial in learning more fine-grained features. We theoretically prove utility improvements in the case that clients with stricter privacy budgets spend budgets unevenly across rounds, compared to clients with more relaxed budgets, who have sufficient budgets to distribute their spend more evenly. Our practical experiments on standard benchmark datasets support our theoretical results and show that, in practice, our algorithms improve the privacy-utility trade-offs compared to baseline schemes.