Parallel Stochastic Newton Method

Mojmir Mutny, Peter Richtarik

Research output: Contribution to journalArticlepeer-review

Abstract

We propose a parallel stochastic Newton method (PSN) for minimizing unconstrained smooth convex functions. We analyze the method in the strongly convex case, and give conditions under which acceleration can be expected when compared to its serial counterpart. We show how PSN can be applied to the empirical risk minimization problem, and demonstrate the practical efficiency of the method through numerical experiments and models of simple matrix classes.
Original languageEnglish
Pages (from-to)404-425
Number of pages22
JournalJournal of Computational Mathematics
Volume36
Issue number3
Early online date29 Mar 2018
DOIs
Publication statusPublished - 2018

Fingerprint

Dive into the research topics of 'Parallel Stochastic Newton Method'. Together they form a unique fingerprint.

Cite this