Differentially Private Bayesian Inference for Exponential Families

By Garrett Bernstein, D. Sheldon. 2018

53 citations recorded by the source corpus.

The study of private inference has been sparked by growing concern regarding the analysis of data when it stems from sensitive sources. We present the first method for private Bayesian inference in exponential families that properly accounts for noise introduced by the privacy mechanism. It is efficient because it works only with sufficient statistics and not individual data. Unlike other methods, it gives properly calibrated posterior beliefs in the non-asymptotic data regime.

Subfields: Artificial Intelligence, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Ashwin Machanavajjhala
2. Michael Hay
3. Gerome Miklau
4. Ryan McKenna
5. D. Sheldon
6. Garrett Bernstein
7. Daniel Kifer
8. Adam D. Smith

In Statistics and Probability

1. D. Sheldon
2. Garrett Bernstein
3. Michael Hay
4. Gerome Miklau
5. Ashwin Machanavajjhala
6. Ryan McKenna

Citations in the corpus

• Fair decision making using privacy-protected data (2019, background): “Estimating posterior distributions from differentially-private statistics has been studied previously [9, 10, 33].”
• Differentially Private Bayesian Linear Regression (2019, methodology): “We leverage a model in which the data and model parameters are latent variables, and noisy sufficient statistics are observed, and then develop MCMC-based techniques to sample from posterior distributions, as done for exponential families in [Bernstein and Sheldon, 2018].”