Winning the NIST Contest: A scalable and general approach to differentially private synthetic data. McKenna, R., Miklau, G., & Sheldon, D. arXiv:2108.04978 [cs], August, 2021. arXiv: 2108.04978
Winning the NIST Contest: A scalable and general approach to differentially private synthetic data [link]Paper  abstract   bibtex   
We propose a general approach for differentially private synthetic data generation, that consists of three steps: (1) select a collection of low-dimensional marginals, (2) measure those marginals with a noise addition mechanism, and (3) generate synthetic data that preserves the measured marginals well. Central to this approach is Private-PGM, a post-processing method that is used to estimate a high-dimensional data distribution from noisy measurements of its marginals. We present two mechanisms, NIST-MST and MST, that are instances of this general approach. NIST-MST was the winning mechanism in the 2018 NIST differential privacy synthetic data competition, and MST is a new mechanism that can work in more general settings, while still performing comparably to NIST-MST. We believe our general approach should be of broad interest, and can be adopted in future mechanisms for synthetic data generation.
@article{mckenna_winning_2021,
	title = {Winning the {NIST} {Contest}: {A} scalable and general approach to differentially private synthetic data},
	shorttitle = {Winning the {NIST} {Contest}},
	url = {http://arxiv.org/abs/2108.04978},
	abstract = {We propose a general approach for differentially private synthetic data generation, that consists of three steps: (1) select a collection of low-dimensional marginals, (2) measure those marginals with a noise addition mechanism, and (3) generate synthetic data that preserves the measured marginals well. Central to this approach is Private-PGM, a post-processing method that is used to estimate a high-dimensional data distribution from noisy measurements of its marginals. We present two mechanisms, NIST-MST and MST, that are instances of this general approach. NIST-MST was the winning mechanism in the 2018 NIST differential privacy synthetic data competition, and MST is a new mechanism that can work in more general settings, while still performing comparably to NIST-MST. We believe our general approach should be of broad interest, and can be adopted in future mechanisms for synthetic data generation.},
	urldate = {2021-08-16},
	journal = {arXiv:2108.04978 [cs]},
	author = {McKenna, Ryan and Miklau, Gerome and Sheldon, Daniel},
	month = aug,
	year = {2021},
	note = {arXiv: 2108.04978},
	keywords = {cryptography, mentions sympy},
}

Downloads: 0