Cookies

We use cookies to ensure that we give you the best experience on our website. By continuing to browse this repository, you give consent for essential cookies to be used. You can read more about our Privacy and Cookie Policy.


Durham Research Online
You are in:

Encrypted accelerated least squares regression.

Esperança, P. M. and Aslett, L. J. M. and Holmes, C. C. (2017) 'Encrypted accelerated least squares regression.', in Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, PMLR 54. Fort Lauderdale, FL, USA: PMLR, pp. 334-343. Proceedings of machine learning research., 54

Abstract

Information that is stored in an encrypted format is, by definition, usually not amenable to statistical analysis or machine learning methods. In this paper we present detailed analysis of coordinate and accelerated gradient descent algorithms which are capable of fitting least squares and penalised ridge regression models, using data encrypted under a fully homomorphic encryption scheme. Gradient descent is shown to dominate in terms of encrypted computational speed, and theoretical results are proven to give parameter bounds which ensure correctness of decryption. The characteristics of encrypted computation are empirically shown to favour a non-standard acceleration technique. This demonstrates the possibility of approximating conventional statistical regression methods using encrypted data without compromising privacy.

Item Type:Book chapter
Full text:(VoR) Version of Record
Download PDF
(406Kb)
Status:Peer-reviewed
Publisher Web site:http://proceedings.mlr.press/v54/esperanca17a/esperanca17a.pdf
Publisher statement:Copyright 2017 by the author(s).
Date accepted:25 January 2017
Date deposited:04 May 2017
Date of first online publication:20 April 2017
Date first made open access:04 May 2017

Save or Share this output

Export:
Export
Look up in GoogleScholar