Privacy with imperfect randomness

Yevgeniy Dodis, Yanqing Yao

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We revisit the impossibility of a variety of cryptographic tasks including privacy and differential privacy with imperfect randomness. For traditional notions of privacy, such as security of encryption, commitment or secret sharing schemes, dramatic impossibility results are known [MP90,DOPS04] for several concrete sources R, including a (seemingly) very “nice and friendly” Santha-Vazirani (SV) source. Somewhat surprisingly, Dodis et al. [DLMV12] showed that non-trivial differential privacy is possible with the SV sources. This suggested a qualitative gap between traditional and differential privacy, and left open the question of whether differential privacy is possible with more realistic (i.e., less structured) sources than the SV sources. Motivated by this question, we introduce a new, modular framework for showing strong impossibility results for (both traditional and differential) privacy under a general imperfect source R. As direct corollaries of our framework, we get the following new results: (1) Existing, but quantitatively improved, impossibility results for traditional privacy, but under a wider variety of sources R. (2) First impossibility results for differential privacy for a variety of realistic sourcesR(including most “block sources”, but not the SV source). (3) Any imperfect source allowing (either traditional or differential) privacy under R admits a certain type of deterministic bit extraction from R.

Original languageEnglish (US)
Title of host publicationAdvances in Cryptology - CRYPTO 2015 - 35th Annual Cryptology Conference, Proceedings
PublisherSpringer Verlag
Pages463-482
Number of pages20
Volume9216
ISBN (Print)9783662479995
DOIs
StatePublished - 2015
Event35th Annual Cryptology Conference, CRYPTO 2015 - Santa Barbara, United States
Duration: Aug 16 2015Aug 20 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9216
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other35th Annual Cryptology Conference, CRYPTO 2015
CountryUnited States
CitySanta Barbara
Period8/16/158/20/15

Fingerprint

Imperfect
Randomness
Cryptography
Privacy
Secret Sharing Scheme
Encryption
Corollary

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Dodis, Y., & Yao, Y. (2015). Privacy with imperfect randomness. In Advances in Cryptology - CRYPTO 2015 - 35th Annual Cryptology Conference, Proceedings (Vol. 9216, pp. 463-482). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9216). Springer Verlag. https://doi.org/10.1007/978-3-662-48000-723

Privacy with imperfect randomness. / Dodis, Yevgeniy; Yao, Yanqing.

Advances in Cryptology - CRYPTO 2015 - 35th Annual Cryptology Conference, Proceedings. Vol. 9216 Springer Verlag, 2015. p. 463-482 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9216).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Dodis, Y & Yao, Y 2015, Privacy with imperfect randomness. in Advances in Cryptology - CRYPTO 2015 - 35th Annual Cryptology Conference, Proceedings. vol. 9216, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9216, Springer Verlag, pp. 463-482, 35th Annual Cryptology Conference, CRYPTO 2015, Santa Barbara, United States, 8/16/15. https://doi.org/10.1007/978-3-662-48000-723
Dodis Y, Yao Y. Privacy with imperfect randomness. In Advances in Cryptology - CRYPTO 2015 - 35th Annual Cryptology Conference, Proceedings. Vol. 9216. Springer Verlag. 2015. p. 463-482. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-662-48000-723
Dodis, Yevgeniy ; Yao, Yanqing. / Privacy with imperfect randomness. Advances in Cryptology - CRYPTO 2015 - 35th Annual Cryptology Conference, Proceedings. Vol. 9216 Springer Verlag, 2015. pp. 463-482 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{c4487c8b50a74861b9619b813db67041,
title = "Privacy with imperfect randomness",
abstract = "We revisit the impossibility of a variety of cryptographic tasks including privacy and differential privacy with imperfect randomness. For traditional notions of privacy, such as security of encryption, commitment or secret sharing schemes, dramatic impossibility results are known [MP90,DOPS04] for several concrete sources R, including a (seemingly) very “nice and friendly” Santha-Vazirani (SV) source. Somewhat surprisingly, Dodis et al. [DLMV12] showed that non-trivial differential privacy is possible with the SV sources. This suggested a qualitative gap between traditional and differential privacy, and left open the question of whether differential privacy is possible with more realistic (i.e., less structured) sources than the SV sources. Motivated by this question, we introduce a new, modular framework for showing strong impossibility results for (both traditional and differential) privacy under a general imperfect source R. As direct corollaries of our framework, we get the following new results: (1) Existing, but quantitatively improved, impossibility results for traditional privacy, but under a wider variety of sources R. (2) First impossibility results for differential privacy for a variety of realistic sourcesR(including most “block sources”, but not the SV source). (3) Any imperfect source allowing (either traditional or differential) privacy under R admits a certain type of deterministic bit extraction from R.",
author = "Yevgeniy Dodis and Yanqing Yao",
year = "2015",
doi = "10.1007/978-3-662-48000-723",
language = "English (US)",
isbn = "9783662479995",
volume = "9216",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "463--482",
booktitle = "Advances in Cryptology - CRYPTO 2015 - 35th Annual Cryptology Conference, Proceedings",

}

TY - GEN

T1 - Privacy with imperfect randomness

AU - Dodis, Yevgeniy

AU - Yao, Yanqing

PY - 2015

Y1 - 2015

N2 - We revisit the impossibility of a variety of cryptographic tasks including privacy and differential privacy with imperfect randomness. For traditional notions of privacy, such as security of encryption, commitment or secret sharing schemes, dramatic impossibility results are known [MP90,DOPS04] for several concrete sources R, including a (seemingly) very “nice and friendly” Santha-Vazirani (SV) source. Somewhat surprisingly, Dodis et al. [DLMV12] showed that non-trivial differential privacy is possible with the SV sources. This suggested a qualitative gap between traditional and differential privacy, and left open the question of whether differential privacy is possible with more realistic (i.e., less structured) sources than the SV sources. Motivated by this question, we introduce a new, modular framework for showing strong impossibility results for (both traditional and differential) privacy under a general imperfect source R. As direct corollaries of our framework, we get the following new results: (1) Existing, but quantitatively improved, impossibility results for traditional privacy, but under a wider variety of sources R. (2) First impossibility results for differential privacy for a variety of realistic sourcesR(including most “block sources”, but not the SV source). (3) Any imperfect source allowing (either traditional or differential) privacy under R admits a certain type of deterministic bit extraction from R.

AB - We revisit the impossibility of a variety of cryptographic tasks including privacy and differential privacy with imperfect randomness. For traditional notions of privacy, such as security of encryption, commitment or secret sharing schemes, dramatic impossibility results are known [MP90,DOPS04] for several concrete sources R, including a (seemingly) very “nice and friendly” Santha-Vazirani (SV) source. Somewhat surprisingly, Dodis et al. [DLMV12] showed that non-trivial differential privacy is possible with the SV sources. This suggested a qualitative gap between traditional and differential privacy, and left open the question of whether differential privacy is possible with more realistic (i.e., less structured) sources than the SV sources. Motivated by this question, we introduce a new, modular framework for showing strong impossibility results for (both traditional and differential) privacy under a general imperfect source R. As direct corollaries of our framework, we get the following new results: (1) Existing, but quantitatively improved, impossibility results for traditional privacy, but under a wider variety of sources R. (2) First impossibility results for differential privacy for a variety of realistic sourcesR(including most “block sources”, but not the SV source). (3) Any imperfect source allowing (either traditional or differential) privacy under R admits a certain type of deterministic bit extraction from R.

UR - http://www.scopus.com/inward/record.url?scp=84943379531&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943379531&partnerID=8YFLogxK

U2 - 10.1007/978-3-662-48000-723

DO - 10.1007/978-3-662-48000-723

M3 - Conference contribution

AN - SCOPUS:84943379531

SN - 9783662479995

VL - 9216

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 463

EP - 482

BT - Advances in Cryptology - CRYPTO 2015 - 35th Annual Cryptology Conference, Proceedings

PB - Springer Verlag

ER -