Subsampled Rényi Differential Privacy and Analytical Moments Accountant

Main Article Content

Yu-Xiang Wang
Borja Balle
Shiva Kasiviswanathan

Abstract

We study the problem of subsampling in differential privacy (DP), a question that is the centerpiece behind many successful differentially private machine learning algorithms. Specifically, we provide a tight upper bound on the Renyi Differential Privacy (RDP) [Mironov, 2017] parameters for algorithms that: (1) subsample the dataset, and then (2) apply a randomized mechanism M to the subsample, in terms of the RDP parameters of M and the subsampling probability parameter.
Our results generalize the moments accounting technique, developed by [Abadi et al. 2016] for the Gaussian mechanism, to any subsampled RDP mechanism.

Article Details

How to Cite
Wang, Yu-Xiang, Borja Balle, and Shiva Kasiviswanathan. 2020. “Subsampled Rényi Differential Privacy and Analytical Moments Accountant”. Journal of Privacy and Confidentiality 10 (2). https://doi.org/10.29012/jpc.723.
Section
TPDP 2018