Acta Mathematica Vietnamica
Search within journal
Volumes and issues
CONVERGENCE OF A MODIFIED CONJUGATE DESCENT METHOD IN THE PRESENCE OF ERRORS
MIN SUNpp. 371-378
Abstract
The conjugate gradient method is a useful method for solving large-scale minimization problems. Errors may arise because of inexact computation. In this paper we study the global convergence of a modified conjugate descent method with strong Wolfe line search in the presence of errors. Preliminary numerical experiments are given to show the efficiency and robustness of the method.