Differentially Private Markov Chain Monte Carlo

Recent developments in differentially private (DP) machine learning and DP Bayesian learning have enabled learning under strong privacy guarantees for the training data subjects. In this paper, we further extend the applicability of DP Bayesian learning by presenting the first general DP Markov chai...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Heikkilä, Mikko A, Jälkö, Joonas, Dikmen, Onur, Honkela, Antti
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 17.06.2019
Subjects
Acceptance tests
Algorithms
Bayesian analysis
Computer simulation
Machine learning
Markov analysis
Markov chains
Monte Carlo simulation
Privacy
Online AccessGet full text
ISSN2331-8422

Cover

More Information
Summary:Recent developments in differentially private (DP) machine learning and DP Bayesian learning have enabled learning under strong privacy guarantees for the training data subjects. In this paper, we further extend the applicability of DP Bayesian learning by presenting the first general DP Markov chain Monte Carlo (MCMC) algorithm whose privacy-guarantees are not subject to unrealistic assumptions on Markov chain convergence and that is applicable to posterior inference in arbitrary models. Our algorithm is based on a decomposition of the Barker acceptance test that allows evaluating the Rényi DP privacy cost of the accept-reject choice. We further show how to improve the DP guarantee through data subsampling and approximate acceptance tests.
Bibliography:content type line 50
SourceType-Working Papers-1
ObjectType-Working Paper/Pre-Print-1
ISSN:2331-8422