2007•Journal of Anyang Institute of TechnologyRequires access

The New Conjugate Gradient Method with Generalized Wolfe Step Size Rule

Jing Shu-jie

Abstract

Conjugate gradient optimization algorithms depend on the search directions with different choice for the parameter in the conjugate gradient directions.In this paper,conditions are given on the parameter to ensure that the conjugate direction is sufficient descent,and a new conjugate gradient method is present.This algorithm only need a smaller memory and to have the better convergence rate.

About this research paper

What this paper is about

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Key concepts: Conjugate gradient method, Derivation of the conjugate gradient method, Conjugate residual method, Conjugate, Nonlinear conjugate gradient method, Gradient descent, Gradient method, Convergence (economics)