点击次数：

论文类型：期刊论文

第一作者：Milzarek, Andre

通讯作者：Milzarek, A (corresponding author), Chinese Univ Hong Kong, Inst Data & Decis Analyt IDDA, Shenzhen Inst Artificial Intelligence & Robot Soc, Shenzhen, Peoples R China.

合写作者：Xiao, Xiantao,Cen, Shicong,Wen, Zaiwen,Ulbrich, Michael

发表时间：2019-01-01

发表刊物：SIAM JOURNAL ON OPTIMIZATION

收录刊物：EI、SCIE

卷号：29

期号：4

页面范围：2916-2948

ISSN号：1052-6234

关键字：nonsmooth stochastic optimization; stochastic approximation; semismooth Newton method; stochastic second order information; global convergence

摘要：In this work, we present a globalized stochastic semismooth Newton method for solving stochastic optimization problems involving smooth nonconvex and nonsmooth convex terms in the objective function. We assume that only noisy gradient and Hessian information of the smooth part of the objective function is available via calling stochastic first and second order oracles. The proposed method can be seen as a hybrid approach combining stochastic semismooth Newton steps and stochastic proximal gradient steps. Two inexact growth conditions are incorporated to monitor the convergence and the acceptance of the semismooth Newton steps and it is shown that the algorithm converges globally to stationary points in expectation and almost surely. We present numerical results and comparisons on l1-regularized logistic regression and nonconvex binary classification that demonstrate the efficiency of the algorithm.