CONVERGENCE PROPERTIES OF CONJUGATEGRADIENT METHODS WITH STRONG WOLFE LINESEARCH  

作　　者：HAN Jiye LIU Guanghui YIN Hongxia(Institute of Applied Mathematics, Academia Silica, Beijing 100080, China) 

出　　处：《Systems Science and Mathematical Sciences》1998年第2期112-116,共5页

摘　　要：In this paper, we investigate the convergence property of the conjugate gradientalgorithms which depend on the search directionsby using different choices for the scalar oh, where 9k is the gradient of f at ac. Undersome assumptions which are slightly weaker than those in [1], we prove a global convergenceresult which allows ac to be set in a wider range than that in [1].In this paper, we investigate the convergence property of the conjugate gradientalgorithms which depend on the search directionsby using different choices for the scalar oh, where 9k is the gradient of f at ac. Undersome assumptions which are slightly weaker than those in [1], we prove a global convergenceresult which allows ac to be set in a wider range than that in [1].

关 键 词：CONJUGATE GRADIENT ALGORITHMS global convergence) UNCONSTRAINED optimization. 

分 类 号：O224[理学—运筹学与控制论]