On the Theory and Practice of Privacy-Preserving Bayesian Data Analysis

James Foulds, Joseph Geumlek, Max Welling, Kamalika Chaudhuri
Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, PMLR R14:118-127, 2016.

Abstract

Bayesian inference has great promise for the privacy-preserving analysis of sensitive data, as posterior sampling automatically preserves differential privacy, an algorithmic notion of data privacy, under certain conditions (Dimitrakakis et al., 2014; Wang et al., 2015). While this one posterior sample (OPS) approach elegantly provides privacy "for free," it is data inefficient in the sense of asymptotic relative efficiency (ARE). We show that a simple alternative based on the Laplace mechanism, the workhorse of differential privacy, is as asymptotically efficient as non-private posterior inference, under general assumptions. This technique also has practical advantages including efficient use of the privacy budget for MCMC. We demonstrate the practicality of our approach on a time-series analysis of sensitive military records from the Afghanistan and Iraq wars disclosed by the Wikileaks organization.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR14-foulds16a, title = {On the Theory and Practice of Privacy-Preserving {B}ayesian Data Analysis}, author = {Foulds, James and Geumlek, Joseph and Welling, Max and Chaudhuri, Kamalika}, booktitle = {Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence}, pages = {118--127}, year = {2016}, editor = {Ihler, Alexander and Janzing, Dominik}, volume = {R14}, series = {Proceedings of Machine Learning Research}, month = {25--29 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r14/main/assets/foulds16a/foulds16a.pdf}, url = {https://proceedings.mlr.press/r14/foulds16a.html}, abstract = {Bayesian inference has great promise for the privacy-preserving analysis of sensitive data, as posterior sampling automatically preserves differential privacy, an algorithmic notion of data privacy, under certain conditions (Dimitrakakis et al., 2014; Wang et al., 2015). While this one posterior sample (OPS) approach elegantly provides privacy "for free," it is data inefficient in the sense of asymptotic relative efficiency (ARE). We show that a simple alternative based on the Laplace mechanism, the workhorse of differential privacy, is as asymptotically efficient as non-private posterior inference, under general assumptions. This technique also has practical advantages including efficient use of the privacy budget for MCMC. We demonstrate the practicality of our approach on a time-series analysis of sensitive military records from the Afghanistan and Iraq wars disclosed by the Wikileaks organization.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T On the Theory and Practice of Privacy-Preserving Bayesian Data Analysis %A James Foulds %A Joseph Geumlek %A Max Welling %A Kamalika Chaudhuri %B Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2016 %E Alexander Ihler %E Dominik Janzing %F pmlr-vR14-foulds16a %I PMLR %P 118--127 %U https://proceedings.mlr.press/r14/foulds16a.html %V R14 %X Bayesian inference has great promise for the privacy-preserving analysis of sensitive data, as posterior sampling automatically preserves differential privacy, an algorithmic notion of data privacy, under certain conditions (Dimitrakakis et al., 2014; Wang et al., 2015). While this one posterior sample (OPS) approach elegantly provides privacy "for free," it is data inefficient in the sense of asymptotic relative efficiency (ARE). We show that a simple alternative based on the Laplace mechanism, the workhorse of differential privacy, is as asymptotically efficient as non-private posterior inference, under general assumptions. This technique also has practical advantages including efficient use of the privacy budget for MCMC. We demonstrate the practicality of our approach on a time-series analysis of sensitive military records from the Afghanistan and Iraq wars disclosed by the Wikileaks organization. %Z Reissued by PMLR on 04 October 2026.
APA
Foulds, J., Geumlek, J., Welling, M. & Chaudhuri, K.. (2016). On the Theory and Practice of Privacy-Preserving Bayesian Data Analysis. Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R14:118-127 Available from https://proceedings.mlr.press/r14/foulds16a.html. Reissued by PMLR on 04 October 2026.

Related Material