Scalable and Secure Logistic Regression via Homomorphic Encryption

被引：111

作者：

Aono, Yoshinori ^{[1
]}

Hayashi, Takuya ^{[1
]}

Le Trieu Phong ^{[1
]}

Wang, Lihua ^{[1
]}

机构：

[1] NICT, Koganei, Tokyo, Japan

来源：

CODASPY'16: PROCEEDINGS OF THE SIXTH ACM CONFERENCE ON DATA AND APPLICATION SECURITY AND PRIVACY | 2016年

关键词：

Logistic regression; additively homomorphic encryption; out sourced computation;

D O I：

10.1145/2857705.2857731

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Logistic regression is a powerful machine learning tool to classify data. When dealing with sensitive data such as private or medical information, cares are necessary. In this paper, we propose a secure system for protecting the training data in logistic regression via homomorphic encryption. Perhaps surprisingly, despite the non-polynomial tasks of training in logistic regression, we show that only additively homomorphic encryption is needed to build our system. Our system is secure and scalable with the dataset size.

引用

页码：142 / 144

页数：3

共 5 条

[1]

[Anonymous], 1978, FDN SEC COMPUT

[2]

Aono Y., 2016, IACR CRYPTOLOGY EPRI

[3] Private predictive analysis on encrypted medical data [J].

Bos, Joppe W. ;

Lauter, Kristin ;

Naehrig, Michael .

JOURNAL OF BIOMEDICAL INFORMATICS, 2014, 50 :234-243

[4]

Lauter K, 2011, PROCEEDINGS OF THE 3RD ACM WORKSHOP CLOUD COMPUTING SECURITY WORKSHOP (CCSW'11), P113

[5] Functional Mechanism: Regression Analysis under Differential Privacy [J].

Zhang, Jun ;

Zhang, Zhenjie ;

Xiao, Xiaokui ;

Yang, Yin ;

Winslett, Marianne .

PROCEEDINGS OF THE VLDB ENDOWMENT, 2012, 5 (11) :1364-1375

← 1 →