Generative Adversarial Privacy

A Data-Driven Approach to Information-Theoretic Privacy

Chong Huang, Peter Kairouz, Lalitha Sankar

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

Abstract

We present a data-driven framework called generative adversarial privacy (GAP). Inspired by recent advancements in generative adversarial networks (GANs), GAP allows the data holder to learn the privatization mechanism directly from the data. Under GAP, finding the optimal privacy mechanism is formulated as a constrained minimax game between a privatizer and an adversary. We show that for appropriately chosen adversarial loss functions, GAP provides privacy guarantees against strong information-theoretic adversaries. We also evaluate GAP's performance on the GENKI face database.

Original languageEnglish (US)
Title of host publicationConference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018
EditorsMichael B. Matthews
PublisherIEEE Computer Society
Pages2162-2166
Number of pages5
ISBN (Electronic)9781538692189
DOIs
StatePublished - Feb 19 2019
Event52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018 - Pacific Grove, United States
Duration: Oct 28 2018Oct 31 2018

Publication series

NameConference Record - Asilomar Conference on Signals, Systems and Computers
Volume2018-October
ISSN (Print)1058-6393

Conference

Conference52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018
CountryUnited States
CityPacific Grove
Period10/28/1810/31/18

Fingerprint

Privatization

Keywords

  • Data Privacy
  • Generative Adversarial Networks
  • Information Theory
  • Minimax Games

ASJC Scopus subject areas

  • Signal Processing
  • Computer Networks and Communications

Cite this

Huang, C., Kairouz, P., & Sankar, L. (2019). Generative Adversarial Privacy: A Data-Driven Approach to Information-Theoretic Privacy. In M. B. Matthews (Ed.), Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018 (pp. 2162-2166). [8645532] (Conference Record - Asilomar Conference on Signals, Systems and Computers; Vol. 2018-October). IEEE Computer Society. https://doi.org/10.1109/ACSSC.2018.8645532

Generative Adversarial Privacy : A Data-Driven Approach to Information-Theoretic Privacy. / Huang, Chong; Kairouz, Peter; Sankar, Lalitha.

Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018. ed. / Michael B. Matthews. IEEE Computer Society, 2019. p. 2162-2166 8645532 (Conference Record - Asilomar Conference on Signals, Systems and Computers; Vol. 2018-October).

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

Huang, C, Kairouz, P & Sankar, L 2019, Generative Adversarial Privacy: A Data-Driven Approach to Information-Theoretic Privacy. in MB Matthews (ed.), Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018., 8645532, Conference Record - Asilomar Conference on Signals, Systems and Computers, vol. 2018-October, IEEE Computer Society, pp. 2162-2166, 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018, Pacific Grove, United States, 10/28/18. https://doi.org/10.1109/ACSSC.2018.8645532
Huang C, Kairouz P, Sankar L. Generative Adversarial Privacy: A Data-Driven Approach to Information-Theoretic Privacy. In Matthews MB, editor, Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018. IEEE Computer Society. 2019. p. 2162-2166. 8645532. (Conference Record - Asilomar Conference on Signals, Systems and Computers). https://doi.org/10.1109/ACSSC.2018.8645532
Huang, Chong ; Kairouz, Peter ; Sankar, Lalitha. / Generative Adversarial Privacy : A Data-Driven Approach to Information-Theoretic Privacy. Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018. editor / Michael B. Matthews. IEEE Computer Society, 2019. pp. 2162-2166 (Conference Record - Asilomar Conference on Signals, Systems and Computers).
@inproceedings{55e7edb9fbc7485a9a1de576142e73e6,
title = "Generative Adversarial Privacy: A Data-Driven Approach to Information-Theoretic Privacy",
abstract = "We present a data-driven framework called generative adversarial privacy (GAP). Inspired by recent advancements in generative adversarial networks (GANs), GAP allows the data holder to learn the privatization mechanism directly from the data. Under GAP, finding the optimal privacy mechanism is formulated as a constrained minimax game between a privatizer and an adversary. We show that for appropriately chosen adversarial loss functions, GAP provides privacy guarantees against strong information-theoretic adversaries. We also evaluate GAP's performance on the GENKI face database.",
keywords = "Data Privacy, Generative Adversarial Networks, Information Theory, Minimax Games",
author = "Chong Huang and Peter Kairouz and Lalitha Sankar",
year = "2019",
month = "2",
day = "19",
doi = "10.1109/ACSSC.2018.8645532",
language = "English (US)",
series = "Conference Record - Asilomar Conference on Signals, Systems and Computers",
publisher = "IEEE Computer Society",
pages = "2162--2166",
editor = "Matthews, {Michael B.}",
booktitle = "Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018",

}

TY - GEN

T1 - Generative Adversarial Privacy

T2 - A Data-Driven Approach to Information-Theoretic Privacy

AU - Huang, Chong

AU - Kairouz, Peter

AU - Sankar, Lalitha

PY - 2019/2/19

Y1 - 2019/2/19

N2 - We present a data-driven framework called generative adversarial privacy (GAP). Inspired by recent advancements in generative adversarial networks (GANs), GAP allows the data holder to learn the privatization mechanism directly from the data. Under GAP, finding the optimal privacy mechanism is formulated as a constrained minimax game between a privatizer and an adversary. We show that for appropriately chosen adversarial loss functions, GAP provides privacy guarantees against strong information-theoretic adversaries. We also evaluate GAP's performance on the GENKI face database.

AB - We present a data-driven framework called generative adversarial privacy (GAP). Inspired by recent advancements in generative adversarial networks (GANs), GAP allows the data holder to learn the privatization mechanism directly from the data. Under GAP, finding the optimal privacy mechanism is formulated as a constrained minimax game between a privatizer and an adversary. We show that for appropriately chosen adversarial loss functions, GAP provides privacy guarantees against strong information-theoretic adversaries. We also evaluate GAP's performance on the GENKI face database.

KW - Data Privacy

KW - Generative Adversarial Networks

KW - Information Theory

KW - Minimax Games

UR - http://www.scopus.com/inward/record.url?scp=85063003069&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063003069&partnerID=8YFLogxK

U2 - 10.1109/ACSSC.2018.8645532

DO - 10.1109/ACSSC.2018.8645532

M3 - Conference contribution

T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers

SP - 2162

EP - 2166

BT - Conference Record of the 52nd Asilomar Conference on Signals, Systems and Computers, ACSSC 2018

A2 - Matthews, Michael B.

PB - IEEE Computer Society

ER -