National Science Library, Chinese Academy of Sciences
  登录 机构网站 ENGLISH
您当前的位置是:首页->详细浏览

期刊名称: Applications of Mathematics
Volume:64    Issue:3        Page:335-350
ISSN:0862-7940

An improved nonmonotone adaptive trust region method期刊论文

作者: Xue Yanqin Liu Hongwei Liu Zexian
DOI:10.21136/AM.2019.0138-18

服务链接:
页码: 335-350
出版者: Springer Berlin Heidelberg,Springer
期刊名称: Applications of Mathematics
ISSN: 0862-7940
卷期: Volume:64    Issue:3
语言: English
摘要: Trust region methods are a class of effective iterative schemes in numerical optimization. In this paper, a new improved nonmonotone adaptive trust region method for solving unconstrained optimization problems is proposed. We construct an approximate model where the approximation to Hessian matrix is updated by the scaled memoryless BFGS update formula, and incorporate a nonmonotone technique with the new proposed adaptive trust region radius. The new ratio to adjusting the next trust region radius is different from the ratio in the traditional trust region methods. Under some suitable and standard assumptions, it is shown that the proposed algorithm possesses global convergence and superlinear convergence. Numerical results demonstrate that the proposed method is very promising.
相关主题: trust region method, nonmonotone technique, Theoretical, Mathematical and Computational Physics, global convergence, Mathematics, Optimization, unconstrained optimization, scaled memoryless BFGS update, 90C30, Classical and Continuum Physics, Analysis, Mathematical and Computational Engineering, Applications of Mathematics, Mathematical optimization, Usage, Mathematical models,

相关文献推荐:

问图书管理员更多图书管理员

学科咨询馆员
学科馆员

电话:
邮件:
问图书馆员

图标说明

在线获取原文 原文传递 详细信息 图书在架状态 图书馆际互借 问图书馆员

常见问题

图书馆开放时间 图书馆位置 借阅要求 您在使用中发现的任何错误,都可以向我们 【报告错误】,非常感谢!

作者信息:×