File(s) under permanent embargo

Generative Adversarial Nets Enhanced Continual Data Release Using Differential Privacy

conference contribution
posted on 2020-01-01, 00:00 authored by Stella HoStella Ho, Youyang Qu, Longxiang GaoLongxiang Gao, Jianxin LiJianxin Li, Yong XiangYong Xiang
© 2020, Springer Nature Switzerland AG. In the era of big data, increasing massive volume of data is generated and published consecutively for both research and commercial purposes. The potential value of sensitive information also attracts interest from adversaries and thereby arises public concern. Current research mostly focuses on privacy-preserving data release in a statistic manner rather than taking the dynamics and correlation of context into consideration. Motivated by this, a novel idea is proposed by combining differential privacy and generative adversarial nets. Generative adversarial nets and its extensions are used to generate a synthetic data set with indistinguishable statistic features while differential privacy guarantees a trade-off between the privacy protection and data utility. Extensive simulation results on real-world data set testify the superiority of the proposed model in terms of privacy protection and improved data utility.

History

Event

Algorithms and Architectures for Parallel Processing. Conference (2019 : 19th : Melbourne, Victoria)

Series

Lecture Notes in Computer Science; 11945

Pagination

418 - 426

Publisher

Springer

Location

Melbourne, Victoria

Place of publication

Cham, Switzerland

Start date

2019-12-09

End date

2019-12-11

ISSN

0302-9743

eISSN

1611-3349

ISBN-13

9783030389604

Language

eng

Publication classification

E1 Full written paper - refereed

Editor/Contributor(s)

S Wen, A Zomaya, L Yang

Title of proceedings

ICA3PP 2019 : Algorithms and architectures for parallel processing : 19th International Conference, ICA3PP 2019, Melbourne, VIC, Australia, December 9-11, 2019, Proceedings