TY - JOUR
T1 - PARALLEL STOCHASTIC NEWTON METHOD
AU - Mutny, Mojmir
AU - Richtarik, Peter
N1 - KAUST Repository Item: Exported on 2021-07-08
PY - 2018
Y1 - 2018
N2 - 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 large quadratic function minimization in general, and empirical risk minimization problems. We demonstrate the practical efficiency of the method through numerical experiments and models of simple matrix classes.
AB - 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 large quadratic function minimization in general, and empirical risk minimization problems. We demonstrate the practical efficiency of the method through numerical experiments and models of simple matrix classes.
UR - http://hdl.handle.net/10754/670054
UR - http://global-sci.org/intro/article_detail/jcm/12268.html
U2 - 10.4208/jcm.1708-m2017-0113
DO - 10.4208/jcm.1708-m2017-0113
M3 - Article
SN - 1991-7139
VL - 36
SP - 404
EP - 425
JO - JOURNAL OF COMPUTATIONAL MATHEMATICS
JF - JOURNAL OF COMPUTATIONAL MATHEMATICS
IS - 3
ER -