Paper 2016/111
Scalable and Secure Logistic Regression via Homomorphic Encryption
Yoshinori Aono, Takuya Hayashi, Le Trieu Phong, and Lihua Wang
Abstract
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 both the training and predicting data in logistic regression via homomorphic encryption. Perhaps surprisingly, despite the non-polynomial tasks of training and predicting in logistic regression, we show that only additively homomorphic encryption is needed to build our system. Indeed, we instantiate our system with Paillier, LWE-based, and ring-LWE-based encryption schemes, highlighting the merits and demerits of each instance. Our system is very scalable in both the dataset size and dimension, tolerating big size for example of hundreds of millions ($10^8$s) records. Besides examining the costs of computation and communication, we carefully test our system over real datasets to demonstrate its accuracies and other related measures such as F-score and AUC.
Metadata
- Available format(s)
-
PDF
- Category
- Applications
- Publication info
- Published elsewhere. Major revision. IEICE Transactions 99-D(8): 2079-2089 (2016)
- DOI
- 10.1587/transinf.2015INP0020
- Keywords
- logistic regressionhomomorphic encryptionPaillierLWEring-LWEoutsourced computationaccuracyF-scorearea under curve
- Contact author(s)
- phong @ nict go jp
- History
- 2017-03-31: revised
- 2016-02-10: received
- See all versions
- Short URL
- https://ia.cr/2016/111
- License
-
CC BY
BibTeX
@misc{cryptoeprint:2016/111,
author = {Yoshinori Aono and Takuya Hayashi and Le Trieu Phong and Lihua Wang},
title = {Scalable and Secure Logistic Regression via Homomorphic Encryption},
howpublished = {Cryptology {ePrint} Archive, Paper 2016/111},
year = {2016},
doi = {10.1587/transinf.2015INP0020},
url = {https://eprint.iacr.org/2016/111}
}
