Cryptology ePrint Archive: Report 2019/425

Homomorphic Training of 30,000 Logistic Regression Models

Flavio Bergamaschi and Shai Halevi and Tzipora T. Halevi and Hamish Hunt

Abstract: In this work, we demonstrate the use the CKKS homomorphic encryption scheme to train a large number of logistic regression models simultaneously, as needed to run a genome-wide association study (GWAS) on encrypted data. Our implementation can train more than 30,000 models (each with four features) in about 20 minutes. To that end, we rely on a similar iterative Nesterov procedure to what was used by Kim, Song, Kim, Lee, and Cheon to train a single model [KSKLC18].

We adapt this method to train many models simultaneously using the SIMD capabilities of the CKKS scheme. We also performed a thorough validation of this iterative method and evaluated its suitability both as a generic method for computing logistic regression models, and specifically for GWAS.

Category / Keywords: applications / Approximate numbers, Homomorphic encryption, GWAS, Implementation, Logistic regression

Original Publication (with minor differences): ACNS 2019

Date: received 24 Apr 2019

Contact author: thalevi at nyu edu

Available format(s): PDF | BibTeX Citation

Version: 20190427:184918 (All versions of this report)

Short URL: ia.cr/2019/425


[ Cryptology ePrint archive ]