Differential privacy and robust statistics in high dimensions

X Liu, W Kong, S Oh - Conference on Learning Theory, 2022 - proceedings.mlr.press
We introduce a universal framework for characterizing the statistical efficiency of a statistical
estimation problem with differential privacy guarantees. Our framework, which we call High …

Privacy for free: Posterior sampling and stochastic gradient monte carlo

YX Wang, S Fienberg, A Smola - … Conference on Machine …, 2015 - proceedings.mlr.press
We consider the problem of Bayesian learning on sensitive datasets and present two simple
but somewhat surprising results that connect Bayesian learning to “differential privacy”, a …

Fast differentially private matrix factorization

Z Liu, YX Wang, A Smola - Proceedings of the 9th ACM Conference on …, 2015 - dl.acm.org
Differentially private collaborative filtering is a challenging task, both in terms of accuracy
and speed. We present a simple algorithm that is provably differentially private, while …

Data-dependent PAC-Bayes priors via differential privacy

GK Dziugaite, DM Roy - Advances in neural information …, 2018 - proceedings.neurips.cc
Abstract The Probably Approximately Correct (PAC) Bayes framework (McAllester, 1999)
can incorporate knowledge about the learning algorithm and (data) distribution through the …

Private-knn: Practical differential privacy for computer vision

Y Zhu, X Yu, M Chandraker… - Proceedings of the IEEE …, 2020 - openaccess.thecvf.com
With increasing ethical and legal concerns on privacy for deep models in visual recognition,
differential privacy has emerged as a mechanism to disguise membership of sensitive data …

Generalized gaussian mechanism for differential privacy

F Liu - IEEE Transactions on Knowledge and Data …, 2018 - ieeexplore.ieee.org
Assessment of disclosure risk is of paramount importance in data privacy research and
applications. The concept of differential privacy (DP) formalizes privacy in probabilistic terms …

Revisiting differentially private linear regression: optimal and adaptive prediction & estimation in unbounded domain

YX Wang - arxiv preprint arxiv:1803.02596, 2018 - arxiv.org
We revisit the problem of linear regression under a differential privacy constraint. By
consolidating existing pieces in the literature, we clarify the correct dependence of the …

Differentially private bayesian linear regression

G Bernstein, DR Sheldon - Advances in Neural Information …, 2019 - proceedings.neurips.cc
Linear regression is an important tool across many fields that work with sensitive human-
sourced data. Significant prior work has focused on producing differentially private point …

On the theory and practice of privacy-preserving Bayesian data analysis

J Foulds, J Geumlek, M Welling… - arxiv preprint arxiv …, 2016 - arxiv.org
Bayesian inference has great promise for the privacy-preserving analysis of sensitive data,
as posterior sampling automatically preserves differential privacy, an algorithmic notion of …

Algorithms for differentially private multi-armed bandits

A Tossou, C Dimitrakakis - Proceedings of the AAAI Conference on …, 2016 - ojs.aaai.org
We present differentially private algorithms for the stochastic Multi-Armed Bandit (MAB)
problem. This is a problem for applications such as adaptive clinical trials, experiment …