Robust and Private Bayesian Inference

Abstract

We examine the robustness and privacy of Bayesian inference, under assumptions on the prior, and with no modifications to the Bayesian framework. First, we generalise the concept of differential privacy to arbitrary dataset distances, outcome spaces and distribution families. We then prove bounds on the robustness of the posterior, introduce a posterior sampling mechanism, show that it is differentially private and provide finite sample bounds for distinguishability-based privacy under a strong adversarial model. Finally, we give examples satisfying our assumptions.

Cite

Text

Dimitrakakis et al. "Robust and Private Bayesian Inference." International Conference on Algorithmic Learning Theory, 2014. doi:10.1007/978-3-319-11662-4_21

Markdown

[Dimitrakakis et al. "Robust and Private Bayesian Inference." International Conference on Algorithmic Learning Theory, 2014.](https://mlanthology.org/alt/2014/dimitrakakis2014alt-robust/) doi:10.1007/978-3-319-11662-4_21

BibTeX

@inproceedings{dimitrakakis2014alt-robust,
  title     = {{Robust and Private Bayesian Inference}},
  author    = {Dimitrakakis, Christos and Nelson, Blaine and Mitrokotsa, Aikaterini and Rubinstein, Benjamin I. P.},
  booktitle = {International Conference on Algorithmic Learning Theory},
  year      = {2014},
  pages     = {291-305},
  doi       = {10.1007/978-3-319-11662-4_21},
  url       = {https://mlanthology.org/alt/2014/dimitrakakis2014alt-robust/}
}