A NONMONOTONE CONJUGATE GRADIENT ALGORITHM FOR UNCONSTRAINED OPTIMIZATION
DAIYuhong
Abstract
DAIYuhong
Abstract
Conjugate gradient methods are very important methods for unconstrained optimization.especially for large scale problems.In this paper,we propose a new conjugate gradient method,in which the technique of nomonotone line search is used.Under mild assumptions.we prove the global convergence of the method.Some numerical results are also presented.
OpenAlex reports 27 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Conjugate gradient methods are very important methods for unconstrained optimization.especially for large scale problems.In this paper,we propose a new conjugate gradient method,in which the technique of nomonotone line search is used.Under mild assumptions.we prove the global convergence of the method.Some numerical results are also presented.
Key concepts: Conjugate gradient method, Nonlinear conjugate gradient method, Conjugate residual method, Derivation of the conjugate gradient method, Convergence (economics), Biconjugate gradient method, Gradient method, Mathematical optimization