带记忆的非单调无约束优化算法的全局收敛性  

Global Convergence of a Memory Gradient Method with Nonmonotone Technique for Unconstrained Optimization

作　　者：陈茜[1] 贺向阳[2] 

机构地区：[1]同济大学应用数学系,200092 [2]上海第二工业大学理学院,201209

出　　处：《上海第二工业大学学报》2007年第3期225-228,共4页Journal of Shanghai Polytechnic University

基　　金：上海市教委科研基金项目(No.05RZ12)

摘　　要：自从非单调线搜索技巧引入非线性优化后,所得的算法得到了成功的应用与扩展。带记忆的梯度方法经常用来求解无约束优化问题,尤其是大规模的问题。将带记忆梯度法与Wolfe非单调线搜索技巧成功融合到一起得到了新算法。证明了该算法全局收敛。The technique of nonmonotone line search has received many successful applications and extensions since it was applied in the nonlinear optimization and the memory gradient method is often used for unconstrained optimization, especially large scale problems. This paper combines the memory gradient method and the nonmonotone wolfe line search and the global convergence is obtained.

关 键 词：无约束优化 记忆梯度法 非单调线搜索 全局收敛性 

分 类 号：O224[理学—运筹学与控制论]