Concentrated Differential Privacy: Simplifications, Extensions, and Lower Bounds

Abstract

"Concentrated differential privacy" was recently introduced by Dwork and Rothblum as a relaxation of differential privacy, which permits sharper analyses of many privacy-preserving computations. We present an alternative formulation of the concept of concentrated differential privacy in terms of the Renyi divergence between the distributions obtained by running an algorithm on neighboring inputs. With this reformulation in hand, we prove sharper quantitative results, establish lower bounds, and raise a few new questions. We also unify this approach with approximate differential privacy by giving an appropriate definition of "approximate concentrated differential privacy."

References in corpus (1)

• Rényi Divergence and Kullback-Leibler Divergence

Cited by in corpus (10)

• How to DP-fy ML: A Practical Guide to Machine Learning with Differential Privacy
• Advanced Probabilistic Couplings for Differential Privacy
• Differentially-Private "Draw and Discard" Machine Learning
• Weighted Distributed Differential Privacy ERM: Convex and Non-convex
• Making Differential Privacy Work for Census Data Users
• Differentially Private Confidence Intervals for Proportions under Stratified Random Sampling
• Unbiased Statistical Estimation and Valid Confidence Intervals Under Differential Privacy
• Privacy-preserving Machine Learning for Medical Image Classification
• Synthetic Tabular Data: Methods, Attacks and Defenses
• Privacy accounting $\epsilon$conomics: Improving differential privacy composition via a posteriori bounds