Cryptology ePrint Archive: Report 2020/272

Privacy-Preserving Fast and Exact Linear Equations Solver with Fully Homomorphic Encryption

Keita Arimitsu and Kazuki Otsuka

Abstract: Privacy and machine learning are difficult to coexist due to their nature: parivacy should be kept from others while machine learning requires large amount of data. Among several possible solutions to this problem, Fully Homomorphic Encryption has been a center of intensive researches in this field. FHE enables linear operations of ciphertext. To take advantage of this property, many protocols to achieve statistical operaions have been proposed. On the other hand, many of them are impractical. Some of the approaches introduce cryptosystems that are not familiar. Moreover, most of their protocols are approximation which might sensitively depend on our choice of parameters. In this paper, we propose fast, simple, and exact privacy-preserving linear equation solver using FHE. Our two-party protocol is secure against at least semi-honest model, and we can exactly calculate the model even without the bootstrapping.

Category / Keywords: applications / fully homomorphic encryption, privacy, machine learning, regression

Date: received 29 Feb 2020

Contact author: keita arimitsu at thinkxinc com,kaz@thinkxinc com

Available format(s): PDF | BibTeX Citation

Version: 20200304:080907 (All versions of this report)

Short URL: ia.cr/2020/272


[ Cryptology ePrint archive ]