Subsampled inexact Newton methods for minimizing large sums of convex functions

DOI10.1093/imanum/drz027zbMath1466.65041arXiv1811.05730OpenAlexW2963952517MaRDI QIDQ5857347

Nataša Krejić, Nataša Krklec Jerinkić, Stefania Bellavia

Publication date: 31 March 2021

Published in: IMA Journal of Numerical Analysis (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1811.05730

zbMATH Keywords

global convergencesuperlinear convergenceinexact Newton methodmean square convergencesubsampled Hessian

Mathematics Subject Classification ID

Numerical optimization and variational techniques (65K10) Methods of quasi-Newton type (90C53)

Related Items (8)

Inexact restoration with subsampled trust-region methods for finite-sum minimization ⋮ Hessian averaging in stochastic Newton methods achieves superlinear convergence ⋮ Spectral projected subgradient method for nonsmooth convex optimization problems ⋮ Unnamed Item ⋮ A generalized worst-case complexity analysis for non-monotone line searches ⋮ Linesearch Newton-CG methods for convex optimization with noise ⋮ An inexact restoration-nonsmooth algorithm with variable accuracy for stochastic nonsmooth convex optimization problems in machine learning and stochastic linear complementarity problems ⋮ LSOS: Line-search second-order stochastic optimization methods for nonconvex finite sums

This page was built for publication: Subsampled inexact Newton methods for minimizing large sums of convex functions