Reproduce “On the Theory and Practice of Privacy-Preserving Bayesian Data Analysis” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

James Foulds, Joseph Geumlek, Max Welling, Kamalika Chaudhuri

The authors present a differentially private Bayesian method that adds calibrated Laplace noise to aggregate sufficient statistics in exponential-family models and reuses one private release for MCMC inference. These asymptotic claims are conditional; war-log performance varied by setting, and inferred Iraq patterns around the troop surge are not causal evidence.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.