Privacy with imperfect randomness

Yevgeniy Dodis, Yanqing Yao

Research output: Contribution to journalArticle

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 ageneral 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 sources R (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.

Fingerprint

Imperfect
Randomness
Cryptography
Privacy
Secret Sharing Scheme
Encryption
Corollary

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

@article{ffe3af94b75547e89d1e28e8d6e61907,
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 ageneral 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 sources R (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 = "10.1007/978-3-662-48000-7_23",
language = "English (US)",
volume = "9216",
pages = "463--482",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

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 ageneral 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 sources R (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 ageneral 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 sources R (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=84959339319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959339319&partnerID=8YFLogxK

U2 - 10.1007/978-3-662-48000-7_23

DO - 10.1007/978-3-662-48000-7_23

M3 - Article

AN - SCOPUS:84959339319

VL - 9216

SP - 463

EP - 482

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -