Private Sampling: A Noiseless Approach for Generating Differentially Private Synthetic Data

DOI10.1137/21M1449944MaRDI QIDQ5100091

March Boedihardjo, Thomas Strohmer, R. V. Vershinin

Publication date: 29 August 2022

Published in: SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2109.14839

zbMATH Keywords

dualityempirical processdifferential privacysynthetic dataBoolean Fourier analysismarginal correction

Mathematics Subject Classification ID

Algorithms in computer science (68Wxx) Probability in computer science (algorithm analysis, random structures, phase transitions, etc.) (68Q87) Computational aspects of data analysis and big data (68T09) Privacy of data (68P27)

Uses Software

• BigGAN

Cites Work

• The cost of privacy: optimal rates of convergence for parameter estimation with differential privacy
• Efficient algorithms for privately releasing marginals via convex relaxations
• The Geometry of Differential Privacy: The Small Database and Approximate Cases
• Faster Algorithms for Privately Releasing Marginals
• On the geometry of differential privacy
• PCPs and the Hardness of Generating Private Synthetic Data
• The Algorithmic Foundations of Differential Privacy
• High-Dimensional Probability
• Minimax Optimal Procedures for Locally Private Estimation
• k-ANONYMITY: A MODEL FOR PROTECTING PRIVACY
• Compressive learning with privacy guarantees
• Analysis of Boolean Functions
• A learning theory approach to noninteractive database privacy
• Theory of Cryptography