An information theoretic approach to post randomization methods under differential privacy

DOI10.1007/S11222-020-09949-3MaRDI QIDQ2209709zbMATH OpenOpenAlexFDO

Authors Fadhel Ayed, Marco Battiston, Federico Camerlenghi

Publication date 4 November 2020

Published in Statistics and Computing (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2009.11257

zbMATH Keywords

categorical variablesdifferential privacydisclosure riskmutual informationpost randomization methods

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Sampling theory, sample surveys (62D05) Linear programming (90C05)

Abstract: Post Randomization Methods (PRAM) are among the most popular disclosure limitation techniques for both categorical and continuous data. In the categorical case, given a stochastic matrix $M$ and a specified variable, an individual belonging to category $i$ is changed to category $j$ with probability $M_{i,j}$. Every approach to choose the randomization matrix $M$ has to balance between two desiderata: 1) preserving as much statistical information from the raw data as possible; 2) guaranteeing the privacy of individuals in the dataset. This trade-off has generally been shown to be very challenging to solve. In this work, we use recent tools from the computer science literature and propose to choose $M$ as the solution of a constrained maximization problems. Specifically, $M$ is chosen as the solution of a constrained maximization problem, where we maximize the Mutual Information between raw and transformed data, given the constraint that the transformation satisfies the notion of Differential Privacy. For the general Categorical model, it is shown how this maximization problem reduces to a convex linear programming and can be therefore solved with known optimization algorithms.

Recommendations

• The post randomisation method for protecting microdata
• Optimality models for PRAM
• A concise theory of randomized response techniques for privacy and confidentiality protection
• \(k\)-anonymous microdata release via post randomisation method
• Universally utility-maximizing privacy mechanisms

Cites work

• A Measure of Disclosure Risk for Microdata
• Assessing identification risk in survey microdata using log-linear models
• Assessing the protection provided by misclassification-based disclosure limitation methods for survey microdata
• Concentrated differential privacy: simplifications, extensions, and lower bounds
• Confidentiality and differential privacy in the dissemination of frequency tables
• Data confidentiality: a review of methods for statistical disclosure limitation and methods for assessing privacy
• Differential privacy for Bayesian inference through posterior sampling
• Differential privacy for functions and functional data
• Elements of statistical disclosure control
• Entropy and information theory.
• Estimating identification disclosure risk using mixed membership models
• Minimax Optimal Procedures for Locally Private Estimation
• On Choosing and Bounding Probability Metrics
• Optimality models for PRAM
• Privacy-preserving statistical estimation with optimal convergence rates
• Randomized response: a survey technique for eliminating evasive answer bias
• The algorithmic foundations of differential privacy
• The post randomisation method for protecting microdata
• Theory of Cryptography

Cited in

(2)

• Data privacy using an evolutionary algorithm for invariant PRAM matrices
• Optimality models for PRAM

Describes a project that uses

Uses Software

• RAPPOR

This page was built for publication: An information theoretic approach to post randomization methods under differential privacy

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2209709)