NEW CONJUGATE GRADIENT METHOD IN UNCONSTRAINED OPTIMIZATION PROBLEM

Abstract

Conjugate gradient method is widely acclaimed to be efficient in solving large scale unconstrained optimization problems. In this paper a new method, constructed based on inexact line search scheme was proposed. The global convergence and conjugate descent property of the new method were established based on some assumptions on the objective function and standard Wolfe conditions. Furthermore, numerical experiments established that the method is highly competitive and effective for the solution of large-scale test functions when subjected to comparison with other methods