Goldstein线搜索下混合共轭梯度法的全局收敛性  

CONVERGENCE PROPERTIES OF A HYBRID CONJUGATE GRADIENT METHODS WITH GOLDSTEIN LINE SEARCH

作　　者：焦宝聪[1] 陈兰平[1] 潘翠英[1] 

机构地区：[1]首都师范大学数学科学学院,北京100037

出　　处：《计算数学》2007年第2期137-146,共10页Mathematica Numerica Sinica

基　　金：国家自然科学基金(60472071);北京市教委科研基金(KM200710028001)资助.

摘　　要：本文结合FR算法和DY算法,给出了一类新的杂交共轭梯度算法,并结合Goldstein线搜索,在较弱的条件下证明了算法的收敛性．数值实验表明了新算法的有效性．In this paper, we propose a hybrid of conjugate gradient methods for unconstrained optimization based on Fletcher-Reeves Algorithm and Dai-Yuan Algorithm, which had taken the advantages of two Algorithms. The convergence of the new methods is proved with the Goldstein line search and without the descent condition. Numerical experiments show that the algorith is efficient.

关 键 词：无约束最优化 非精确线搜索 共轭梯度法 全局收敛性 

分 类 号：O224[理学—运筹学与控制论]