Enregistré dans:
Détails bibliographiques
Auteurs principaux: Lin, Yingyu, Ma, Yi-An, Wang, Yu-Xiang, Redberg, Rachel, Bu, Zhiqi
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:https://arxiv.org/abs/2310.14661
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913337409798144
author Lin, Yingyu
Ma, Yi-An
Wang, Yu-Xiang
Redberg, Rachel
Bu, Zhiqi
author_facet Lin, Yingyu
Ma, Yi-An
Wang, Yu-Xiang
Redberg, Rachel
Bu, Zhiqi
contents Posterior sampling, i.e., exponential mechanism to sample from the posterior distribution, provides $\varepsilon$-pure differential privacy (DP) guarantees and does not suffer from potentially unbounded privacy breach introduced by $(\varepsilon,δ)$-approximate DP. In practice, however, one needs to apply approximate sampling methods such as Markov chain Monte Carlo (MCMC), thus re-introducing the unappealing $δ$-approximation error into the privacy guarantees. To bridge this gap, we propose the Approximate SAample Perturbation (abbr. ASAP) algorithm which perturbs an MCMC sample with noise proportional to its Wasserstein-infinity ($W_\infty$) distance from a reference distribution that satisfies pure DP or pure Gaussian DP (i.e., $δ=0$). We then leverage a Metropolis-Hastings algorithm to generate the sample and prove that the algorithm converges in $W_\infty$ distance. We show that by combining our new techniques with a localization step, we obtain the first nearly linear-time algorithm that achieves the optimal rates in the DP-ERM problem with strongly convex and smooth losses.
format Preprint
id arxiv_https___arxiv_org_abs_2310_14661
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Tractable MCMC for Private Learning with Pure and Gaussian Differential Privacy
Lin, Yingyu
Ma, Yi-An
Wang, Yu-Xiang
Redberg, Rachel
Bu, Zhiqi
Machine Learning
Posterior sampling, i.e., exponential mechanism to sample from the posterior distribution, provides $\varepsilon$-pure differential privacy (DP) guarantees and does not suffer from potentially unbounded privacy breach introduced by $(\varepsilon,δ)$-approximate DP. In practice, however, one needs to apply approximate sampling methods such as Markov chain Monte Carlo (MCMC), thus re-introducing the unappealing $δ$-approximation error into the privacy guarantees. To bridge this gap, we propose the Approximate SAample Perturbation (abbr. ASAP) algorithm which perturbs an MCMC sample with noise proportional to its Wasserstein-infinity ($W_\infty$) distance from a reference distribution that satisfies pure DP or pure Gaussian DP (i.e., $δ=0$). We then leverage a Metropolis-Hastings algorithm to generate the sample and prove that the algorithm converges in $W_\infty$ distance. We show that by combining our new techniques with a localization step, we obtain the first nearly linear-time algorithm that achieves the optimal rates in the DP-ERM problem with strongly convex and smooth losses.
title Tractable MCMC for Private Learning with Pure and Gaussian Differential Privacy
topic Machine Learning
url https://arxiv.org/abs/2310.14661