Newton-GSOR method for solving large-scale unconstrained optimization problems
Abstract
Unconstrained convex optimization problems have enormous applications in various field of science and engineering. Different iterative methods are available in literature to solve such problem, and Newton method is among the oldest and simplest one. Due to slow convergence rate of Newton's methods, many research have been carried out to modify the Newton's method for faster convergence rate. In 2019, Ghazali et al. modified Newton's method and proposed Netwon-SOR method, which is a combination of Newton method with SOR iterative method to solve a linear system. In this paper, we propose a modification of Newton-SOR method by modifying SOR method to generalized SOR method. Numerical experiments are carried out to check the efficiently of the proposed method.
Cite
@article{arxiv.2311.05311,
title = {Newton-GSOR method for solving large-scale unconstrained optimization problems},
author = {Santoshi Subhalaxmi Ray and Manideepa Saha},
journal= {arXiv preprint arXiv:2311.05311},
year = {2023}
}