AIM

Author:

McKenna Ryan1,Mullins Brett1,Sheldon Daniel1,Miklau Gerome1

Affiliation:

1. University of Massachusetts, Amherst, Massachusetts

Abstract

We propose AIM, a new algorithm for differentially private synthetic data generation. AIM is a workload-adaptive algorithm within the paradigm of algorithms that first selects a set of queries, then privately measures those queries, and finally generates synthetic data from the noisy measurements. It uses a set of innovative features to iteratively select the most useful measurements, reflecting both their relevance to the workload and their value in approximating the input data. We also provide analytic expressions to bound per-query error with high probability which can be used to construct confidence intervals and inform users about the accuracy of generated data. We show empirically that AIM consistently outperforms a wide variety of existing mechanisms across a variety of experimental settings.

Publisher

Association for Computing Machinery (ACM)

Subject

General Earth and Planetary Sciences,Water Science and Technology,Geography, Planning and Development

Reference57 articles.

1. Privacy Preserving Synthetic Data Release Using Deep Learning

2. Hassan Jameel Asghar , Ming Ding , Thierry Rakotoarivelo , Sirine Mrabet , and Mohamed Ali Kâafar . 2019. Differentially Private Release of High-Dimensional Datasets using the Gaussian Copula. CoRR abs/1902.01499 ( 2019 ). arXiv:1902.01499 http://arxiv.org/abs/1902.01499 Hassan Jameel Asghar, Ming Ding, Thierry Rakotoarivelo, Sirine Mrabet, and Mohamed Ali Kâafar. 2019. Differentially Private Release of High-Dimensional Datasets using the Gaussian Copula. CoRR abs/1902.01499 (2019). arXiv:1902.01499 http://arxiv.org/abs/1902.01499

3. Sergul Aydore , William Brown , Michael Kearns , Krishnaram Kenthapadi , Luca Melis , Aaron Roth , and Ankit A Siva . 2021 . Differentially Private Query Release Through Adaptive Projection . In Proceedings of the 38th International Conference on Machine Learning (Proceedings of Machine Learning Research), Marina Meila and Tong Zhang (Eds.) , Vol. 139 . PMLR, 457--467. https://proceedings.mlr.press/v139/aydore21a.html Sergul Aydore, William Brown, Michael Kearns, Krishnaram Kenthapadi, Luca Melis, Aaron Roth, and Ankit A Siva. 2021. Differentially Private Query Release Through Adaptive Projection. In Proceedings of the 38th International Conference on Machine Learning (Proceedings of Machine Learning Research), Marina Meila and Tong Zhang (Eds.), Vol. 139. PMLR, 457--467. https://proceedings.mlr.press/v139/aydore21a.html

4. Plausible deniability for privacy-preserving data synthesis

5. Concentrated Differential Privacy: Simplifications, Extensions, and Lower Bounds

Cited by 7 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Does Differentially Private Synthetic Data Lead to Synthetic Discoveries?;Methods of Information in Medicine;2024-08-13

2. Performance Truthfulness of Differential Privacy for DB Testing;Proceedings of the Tenth International Workshop on Testing Database Systems;2024-06-09

3. Continual Release of Differentially Private Synthetic Data from Longitudinal Data Collections;Proceedings of the ACM on Management of Data;2024-05-10

4. Differentially Private Data Generation with Missing Data;Proceedings of the VLDB Endowment;2024-04

5. Benchmarking Evaluation Protocols for Classifiers Trained on Differentially Private Synthetic Data;IEEE Access;2024

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3