Filters








5 Hits in 0.93 sec

DPSyn: Experiences in the NIST Differential Privacy Data Synthesis Challenges [article]

Ninghui Li and Zhikun Zhang and Tianhao Wang
2021 arXiv   pre-print
We summarize the experience of participating in two differential privacy competitions organized by the National Institute of Standards and Technology (NIST).  ...  partially funded by NSFC under grant No. 61731004, U1911401, Alibaba-Zhejiang University Joint Research Institute of Frontier Technologies, the Helmholtz Association within the project "Trustworthy Federated Data  ...  the data synthesis) .  ... 
arXiv:2106.12949v1 fatcat:lv64al5tdbeczpre5j2wqmyifm

Privacy-Preserving Collaborative Data Collection and Analysis with Many Missing Values

Yuichi Sei, Andrew J, Hiroshi Okumura, Akihiko Ohsuga
2022 IEEE Transactions on Dependable and Secure Computing  
Using differential privacy (the de facto standard) as a privacy metric, we conduct experiments on synthetic and real data, including COVID-19related data.  ...  The patient data are anonymized and sent to a data collection server.  ...  Differentially Private Synthetic Datasets Generation The literature includes several studies on differentially private data synthesis, such as [38] - [42] .  ... 
doi:10.1109/tdsc.2022.3174887 fatcat:v4p6elzep5dvtc5wqrimgcqtfu

DPSyn: Experiences in the NIST Differential Privacy Data Synthesis Challenges

Tianhao Wang, Ninghui Li, Zhikun Zhang
2021 Journal of Privacy and Confidentiality  
We summarize the experience of participating in two differential privacycompetitions organized by the National Institute of Standards and Technology (NIST).  ...  partially funded by NSFC under grant No. 61731004, U1911401, Alibaba-Zhejiang University Joint Research Institute of Frontier Technologies, the Helmholtz Association within the project "Trustworthy Federated Data  ...  the data synthesis) .  ... 
doi:10.29012/jpc.775 fatcat:kpfhwecbbfbdzj6qfwhhgwjb44

Winning the NIST Contest: A scalable and general approach to differentially private synthetic data

Ryan McKenna, Gerome Miklau, Daniel Sheldon
2021 Journal of Privacy and Confidentiality  
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  ...  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.  ...  Privsyn: Differentially private data synthesis. In 30th USENIX Security Symposium (USENIX Security 21).  ... 
doi:10.29012/jpc.778 fatcat:b2s37gulojbxxm2buyrfzw7vq4

HDPView: Differentially Private Materialized View for Exploring High Dimensional Relational Data [article]

Fumiyuki Kato, Tsubasa Takahashi, Shun Takagi, Yang Cao, Seng Pei Liew, Masatoshi Yoshikawa
2022 arXiv   pre-print
To solve the above issues, we propose HDPView, which creates a differentially private materialized view by well-designed recursive bisected partitioning on an original data cube, i.e., count tensor.  ...  Our method searches for block partitioning to minimize the error for the counting query, in addition to randomizing the convergence, by choosing the effective cutting points in a differentially private  ...  Private data synthesis.  ... 
arXiv:2203.06791v2 fatcat:24yczflhzzctpahgsrfnhmpmey