Differentially Private Federated Bayesian Optimization with Distributed Exploration

TL;DR

This paper introduces a differentially private federated Bayesian optimization algorithm that ensures user privacy while maintaining high utility, leveraging distributed exploration and theoretical privacy-utility guarantees.

Contribution

It develops DP-FTS-DE, a novel algorithm integrating differential privacy into federated Thompson sampling with distributed exploration, providing theoretical privacy and utility guarantees.

Findings

Achieves high utility with strong privacy guarantees.

Demonstrates effective privacy-utility trade-off.

Performs competitively on real-world tasks.

Abstract

Bayesian optimization (BO) has recently been extended to the federated learning (FL) setting by the federated Thompson sampling (FTS) algorithm, which has promising applications such as federated hyperparameter tuning. However, FTS is not equipped with a rigorous privacy guarantee which is an important consideration in FL. Recent works have incorporated differential privacy (DP) into the training of deep neural networks through a general framework for adding DP to iterative algorithms. Following this general DP framework, our work here integrates DP into FTS to preserve user-level privacy. We also leverage the ability of this general DP framework to handle different parameter vectors, as well as the technique of local modeling for BO, to further improve the utility of our algorithm through distributed exploration (DE). The resulting differentially private FTS with DE (DP-FTS-DE) algorithm is endowed with theoretical guarantees for both the privacy and utility and is amenable to interesting theoretical insights about the privacy-utility trade-off. We also use real-world experiments to show that DP-FTS-DE achieves high utility (competitive performance) with a strong privacy guarantee (small privacy loss) and induces a trade-off between privacy and utility.