ITLGMLFeb 2, 2021

Local Differential Privacy Is Equivalent to Contraction of $E_γ$-Divergence

arXiv:2102.01258v139 citations
AI Analysis

This work provides a new theoretical framework for understanding LDP, which is significant for researchers working on privacy-preserving data analysis.

This paper redefines Local Differential Privacy (LDP) constraints using the contraction coefficient of the E_γ-divergence, and then generalizes this to arbitrary f-divergences. This new formulation enables the study of privacy-utility trade-offs in various statistical problems, including testing and estimation.

We investigate the local differential privacy (LDP) guarantees of a randomized privacy mechanism via its contraction properties. We first show that LDP constraints can be equivalently cast in terms of the contraction coefficient of the $E_γ$-divergence. We then use this equivalent formula to express LDP guarantees of privacy mechanisms in terms of contraction coefficients of arbitrary $f$-divergences. When combined with standard estimation-theoretic tools (such as Le Cam's and Fano's converse methods), this result allows us to study the trade-off between privacy and utility in several testing and minimax and Bayesian estimation problems.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes