Composable and versatile privacy via truncated CDP

Mark Bun; Guy N. Rothblum; Cynthia Dwork; Thomas Steinke

doi:10.1145/3188745.3188946

Composable and versatile privacy via truncated CDP

Mark Bun, Guy N. Rothblum, Cynthia Dwork, Thomas Steinke

Department of Computer Science and Applied Mathematics

Weizmann Institute of Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We propose truncated concentrated differential privacy (tCDP), a refinement of differential privacy and of concentrated differential privacy. This new definition provides robust and efficient composition guarantees, supports powerful algorithmic techniques such as privacy amplification via sub-sampling, and enables more accurate statistical analyses. In particular, we show a central task for which the new definition enables exponential accuracy improvement.

Original language	English
Title of host publication	STOC 2018 - Proceedings of the 50th Annual ACM SIGACT Symposium on Theory of Computing
Editors	Monika Henzinger, David Kempe, Ilias Diakonikolas
Pages	74-86
Number of pages	13
ISBN (Electronic)	9781450355599
DOIs	https://doi.org/10.1145/3188745.3188946
State	Published - Jun 2018
Event	50th Annual ACM Symposium on Theory of Computing - United States, CA, Los Angeles Duration: 25 Jun 2018 → 29 Jun 2018 Conference number: 50th

Publication series

Name	Proceedings of the Annual ACM Symposium on Theory of Computing
ISSN (Print)	0737-8017

Conference

Conference	50th Annual ACM Symposium on Theory of Computing
Abbreviated title	STOC 2018
Period	25/06/18 → 29/06/18

All Science Journal Classification (ASJC) codes

Software

Access to Document

10.1145/3188745.3188946

Cite this

@inproceedings{38da612a26094a1a98c834c5a5347c7b,

title = "Composable and versatile privacy via truncated CDP",

abstract = "We propose truncated concentrated differential privacy (tCDP), a refinement of differential privacy and of concentrated differential privacy. This new definition provides robust and efficient composition guarantees, supports powerful algorithmic techniques such as privacy amplification via sub-sampling, and enables more accurate statistical analyses. In particular, we show a central task for which the new definition enables exponential accuracy improvement.",

author = "Mark Bun and Rothblum, \{Guy N.\} and Cynthia Dwork and Thomas Steinke",

note = "We thank the Isaac Newton Institute for Mathematical Sciences, Cambridge U.K., for support and hospitality during the programme “Data Linkage and Anonymisation” where part of this work was done under the support of EPSRC grant no EP/K032208/1. G.N.R.{\textquoteright}s research supported by the ISRAEL SCIENCE FOUNDATION (grant No. 5219/17); 50th Annual ACM Symposium on Theory of Computing ; Conference date: 25-06-2018 Through 29-06-2018",

year = "2018",

month = jun,

doi = "10.1145/3188745.3188946",

language = "الإنجليزيّة",

series = "Proceedings of the Annual ACM Symposium on Theory of Computing",

pages = "74--86",

editor = "Monika Henzinger and David Kempe and Ilias Diakonikolas",

booktitle = "STOC 2018 - Proceedings of the 50th Annual ACM SIGACT Symposium on Theory of Computing",

}

Bun, M, Rothblum, GN, Dwork, C & Steinke, T 2018, Composable and versatile privacy via truncated CDP. in M Henzinger, D Kempe & I Diakonikolas (eds), STOC 2018 - Proceedings of the 50th Annual ACM SIGACT Symposium on Theory of Computing. Proceedings of the Annual ACM Symposium on Theory of Computing, pp. 74-86, 50th Annual ACM Symposium on Theory of Computing, 25/06/18. https://doi.org/10.1145/3188745.3188946

Composable and versatile privacy via truncated CDP. / Bun, Mark; Rothblum, Guy N.; Dwork, Cynthia et al.
STOC 2018 - Proceedings of the 50th Annual ACM SIGACT Symposium on Theory of Computing. ed. / Monika Henzinger; David Kempe; Ilias Diakonikolas. 2018. p. 74-86 (Proceedings of the Annual ACM Symposium on Theory of Computing).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Composable and versatile privacy via truncated CDP

AU - Bun, Mark

AU - Rothblum, Guy N.

AU - Dwork, Cynthia

AU - Steinke, Thomas

N1 - Conference code: 50th

PY - 2018/6

Y1 - 2018/6

N2 - We propose truncated concentrated differential privacy (tCDP), a refinement of differential privacy and of concentrated differential privacy. This new definition provides robust and efficient composition guarantees, supports powerful algorithmic techniques such as privacy amplification via sub-sampling, and enables more accurate statistical analyses. In particular, we show a central task for which the new definition enables exponential accuracy improvement.

AB - We propose truncated concentrated differential privacy (tCDP), a refinement of differential privacy and of concentrated differential privacy. This new definition provides robust and efficient composition guarantees, supports powerful algorithmic techniques such as privacy amplification via sub-sampling, and enables more accurate statistical analyses. In particular, we show a central task for which the new definition enables exponential accuracy improvement.

UR - http://www.scopus.com/inward/record.url?scp=85049884330&partnerID=8YFLogxK

U2 - 10.1145/3188745.3188946

DO - 10.1145/3188745.3188946

M3 - منشور من مؤتمر

T3 - Proceedings of the Annual ACM Symposium on Theory of Computing

SP - 74

EP - 86

BT - STOC 2018 - Proceedings of the 50th Annual ACM SIGACT Symposium on Theory of Computing

A2 - Henzinger, Monika

A2 - Kempe, David

A2 - Diakonikolas, Ilias

T2 - 50th Annual ACM Symposium on Theory of Computing

Y2 - 25 June 2018 through 29 June 2018

ER -

Composable and versatile privacy via truncated CDP

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this