Computer Science > Machine Learning
[Submitted on 21 May 2021 (this version), latest version 18 Oct 2021 (v2)]
Title:Privacy Amplification Via Bernoulli Sampling
View PDFAbstract:Balancing privacy and accuracy is a major challenge in designing differentially private machine learning algorithms. To improve this tradeoff, prior work has looked at privacy amplification methods which analyze how common training operations such as iteration and subsampling the data can lead to higher privacy. In this paper, we analyze privacy amplification properties of a new operation, sampling from the posterior, that is used in Bayesian inference. In particular, we look at Bernoulli sampling from a posterior that is described by a differentially private parameter. We provide an algorithm to compute the amplification factor in this setting, and establish upper and lower bounds on this factor. Finally, we look at what happens when we draw k posterior samples instead of one.
Submission history
From: Jacob Imola [view email][v1] Fri, 21 May 2021 22:34:32 UTC (106 KB)
[v2] Mon, 18 Oct 2021 21:09:38 UTC (131 KB)
Current browse context:
cs.LG
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
IArxiv Recommender
(What is IArxiv?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.