Using robust estimation theory to design efficient secure multiparty linear regression

Fida K. Dankar, Sabri Boughorbel, Radja Badji

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Various biomedical research studies, such as large-population studies and studies on rare diseases, require sharing of data across multiple sources or institutions. In fact, data sharing will enable the collection of more cases for analysis and thus increase the statistical power of the study. However, combining data from various sources poses privacy risks. A number of protocols have been proposed in the literature to address the privacy concerns; but these protocols do not fully deliver either on privacy or complexity. The main reason lies in the methodology used to design these secure algorithms. It is based on translating regular algorithms into secure versions using cryptographic procedures and tricks rather than on establishing robust theory for designing secure and communication free distributed algorithms. In this paper, we use well-established theoretical results to design a secure and low communication linear regression protocol. The method used is comprehensive and can be generalized to other estimators.

Original languageEnglish
JournalCEUR Workshop Proceedings
Volume1558
Publication statusPublished - 2016

Fingerprint

Linear regression
Network protocols
Communication
Parallel algorithms

Keywords

  • Data sharing
  • Information privacy
  • Linear regression
  • Secure multiparty computation

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Using robust estimation theory to design efficient secure multiparty linear regression. / Dankar, Fida K.; Boughorbel, Sabri; Badji, Radja.

In: CEUR Workshop Proceedings, Vol. 1558, 2016.

Research output: Contribution to journalArticle

@article{82da62a0c77c4658abc3ebc7120ff8d2,
title = "Using robust estimation theory to design efficient secure multiparty linear regression",
abstract = "Various biomedical research studies, such as large-population studies and studies on rare diseases, require sharing of data across multiple sources or institutions. In fact, data sharing will enable the collection of more cases for analysis and thus increase the statistical power of the study. However, combining data from various sources poses privacy risks. A number of protocols have been proposed in the literature to address the privacy concerns; but these protocols do not fully deliver either on privacy or complexity. The main reason lies in the methodology used to design these secure algorithms. It is based on translating regular algorithms into secure versions using cryptographic procedures and tricks rather than on establishing robust theory for designing secure and communication free distributed algorithms. In this paper, we use well-established theoretical results to design a secure and low communication linear regression protocol. The method used is comprehensive and can be generalized to other estimators.",
keywords = "Data sharing, Information privacy, Linear regression, Secure multiparty computation",
author = "Dankar, {Fida K.} and Sabri Boughorbel and Radja Badji",
year = "2016",
language = "English",
volume = "1558",
journal = "CEUR Workshop Proceedings",
issn = "1613-0073",
publisher = "CEUR-WS",

}

TY - JOUR

T1 - Using robust estimation theory to design efficient secure multiparty linear regression

AU - Dankar, Fida K.

AU - Boughorbel, Sabri

AU - Badji, Radja

PY - 2016

Y1 - 2016

N2 - Various biomedical research studies, such as large-population studies and studies on rare diseases, require sharing of data across multiple sources or institutions. In fact, data sharing will enable the collection of more cases for analysis and thus increase the statistical power of the study. However, combining data from various sources poses privacy risks. A number of protocols have been proposed in the literature to address the privacy concerns; but these protocols do not fully deliver either on privacy or complexity. The main reason lies in the methodology used to design these secure algorithms. It is based on translating regular algorithms into secure versions using cryptographic procedures and tricks rather than on establishing robust theory for designing secure and communication free distributed algorithms. In this paper, we use well-established theoretical results to design a secure and low communication linear regression protocol. The method used is comprehensive and can be generalized to other estimators.

AB - Various biomedical research studies, such as large-population studies and studies on rare diseases, require sharing of data across multiple sources or institutions. In fact, data sharing will enable the collection of more cases for analysis and thus increase the statistical power of the study. However, combining data from various sources poses privacy risks. A number of protocols have been proposed in the literature to address the privacy concerns; but these protocols do not fully deliver either on privacy or complexity. The main reason lies in the methodology used to design these secure algorithms. It is based on translating regular algorithms into secure versions using cryptographic procedures and tricks rather than on establishing robust theory for designing secure and communication free distributed algorithms. In this paper, we use well-established theoretical results to design a secure and low communication linear regression protocol. The method used is comprehensive and can be generalized to other estimators.

KW - Data sharing

KW - Information privacy

KW - Linear regression

KW - Secure multiparty computation

UR - http://www.scopus.com/inward/record.url?scp=84964555358&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964555358&partnerID=8YFLogxK

M3 - Article

VL - 1558

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

SN - 1613-0073

ER -