Numerical Composition of Differential Privacy
Journal of Privacy and Confidentiality · 11.02.2024 · 14 (1)
DOI: 10.29012/jpc.870 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Privacy and Confidentiality, online erschienen am 11.02.2024. Verfasst von Sivakanth Gopi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 31-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
We give a fast algorithm to optimally compose privacy guarantees of differentially private (DP) algorithms to arbitrary accuracy. Our method is based on the notion of \emph{privacy loss random variables} to quantify the privacy loss of DP algorithms.The running time and memory needed for our algorithm to approximate the privacy curve of a DP algorithm composed with itself $k$ times is $\tilde{O}(\sqrt{k})$. This improves over the best prior method by Koskela et al. (2021) which requires $\tilde{\Omega}(k^{1.5})$ running time. We demonstrate the utility of our algorithm by accurately computing the privacy loss of DP-SGD algorithm of Abadi et al. (2016) and showing that our algorithm speeds up the privacy computations by a few orders of magnitude compared to prior work, while maintaining similar accuracy.
Zitieren
APA
Gopi, S., Lee, Y. T., & Wutschitz, L. (2024). Numerical Composition of Differential Privacy. Journal of Privacy and Confidentiality, 14(1). https://doi.org/10.29012/jpc.870
Vancouver
Gopi S, Lee YT, Wutschitz L. Numerical Composition of Differential Privacy. Journal of Privacy and Confidentiality. 2024;14(1). doi:10.29012/jpc.870
BibTeX
@article{Gopi2024,
author = {Gopi, Sivakanth and Lee, Yin Tat and Wutschitz, Lukas},
title = {Numerical Composition of Differential Privacy},
journal = {Journal of Privacy and Confidentiality},
year = {2024},
volume = {14},
number = {1},
doi = {10.29012/jpc.870}
}
EUVIMED