期刊文献+

修改Broyden非凸族在一般Wolfe搜索下的收敛性 被引量:5

CONVERGENCE PROPERTIES OF THE PRECONVEX PART OF MODIFIED BROYDEN'S FAMILY WITH A GENERALIZED WOLFE LINESEARCHES
原文传递
导出
摘要 近来,韦等提出了一类新的拟牛顿方程Bk+1Sk=yk^*=yk+AkSk,Ak为一矩阵,并在此基础上给出了两种类型的修改Broyden族(MBC).作者利用一般Wolfe搜索技术,与修改Broyden族相结合,证明了在适当的条件下修改Broyden非凸族具有全局收敛性和超线性收敛速度. Recently,Wei and other authors proposed a class of new quasi-Newton equa- tions B_(k+1)s_k=y_l~*=y_k+A_ks_k,where A_k is some matrix,and based on these,they gave two kinds of modified Broyden's familes(MBC).In this paper,generalized Wolfe linesearches proce- dures are used,which are combined with the modified Broyden's family.Under some suitable conditions,we prove the global and superlinear convergence property of the preconvex part of the modified Broyden's family.
出处 《系统科学与数学》 CSCD 北大核心 2007年第2期194-207,共14页 Journal of Systems Science and Mathematical Sciences
基金 国家自然科学基金(10161002) 广西自然科学基金(0135004)资助项目.
关键词 Broyden非凸族 无约束最优化 全局收敛性 超线性收敛性. Preconvex part of Broyden's family, unconstrained optimization, globalconvergence, superlinear convergence.
  • 相关文献

参考文献13

  • 1Wei Z,Yu G.Some recent progress in unconstrained nonlinear optimization.in Proceedings of the 2003's International Conference on Numerical Optimization and Numerical Linear Algebra (Yuan Yaxiang eds.).Science Beijing/New York,2004,110-141.
  • 2王宜举,修乃华.非线性规划理论和算法.西安:陕西科学技术出版社,2004.
  • 3Wei Z,Yu G,Yuan G and Lian G.The superlinear convergence of a modified BFGS type method for Unconstrained Optimization.Computational Optimization and Applications,2004,25:315-332.
  • 4Wei Z,Xiao Y.Global and superlinear convergence of a new Broyden class of quasin-Newton methods.College of Mathematics and Information Sciences,Guangxi university,Nanning,Guangxi,P.R.China,2003.
  • 5Liu G,Han J and Xu Z.Global convergence of the variable metric algorithms with a generalized wolfe linesearch.Journal of Mathematical Research and Exposition,1995,44:99-508.
  • 6袁亚湘 孙文渝.最优化理论和方法[M].科学技术出版社,1997..
  • 7Byrd R and Nocedal J.A tool for the analysis of quasi-Newton methods with application to unconstrained minimization.SIAM Journal on Numerical Analysis,1989,26:727-739.
  • 8Byrd R,Nocedal J and Yuan Y.Global convergence of a class of quasi-Newton methods on convex problems.SIAM Journal on Numerical Analysis,1987,24:1171-1189.
  • 9Zhang Y and Tewarson R P.Quasi-Newton algorithms with updates from the preconvex part of broyden's family.IAM Journal of Numerical Analysis,1988,8:487-509.
  • 10韩继业,刘光辉.无约束最优化线搜索一般模型及BFGS方法的整体收敛性[J].应用数学学报,1995,18(1):112-122. 被引量:20

二级参考文献3

  • 1Zhang Y,IMA J Numer Anal,1988年,8卷,487页
  • 2彭积明,计算数学,1994年,16卷,204页
  • 3袁亚湘,计算数学,1994年,16卷,102页

共引文献47

同被引文献13

  • 1Li D H,Fukushima M. A Modified BFGS Method and Its Global Convergence in Nonconvex Minimization [J]. J Comput App Math ,2001,129: 15-35.
  • 2Wei Z X,I.i G Y,Qi I. Q. New Quasi-Newton Methods for Unconstrained Optimization Problems [J]. Appl Math Comput ,2006,175: 1156-1188.
  • 3Gu N Z, Mo J T. Incorporating Nonmonotone Strategies into the Trust Region Method for UnconstrainedOptimization[J]. Comput Math Appl , 2008,55: 2158- 2172.
  • 4Wei Z X, Yu G H, Yuan G l.,etal. The Superlinear Convergence of a Modified BFGS Type Method for Unconstrained Optimization [J]. Comput Optim Appl, 2004,29 : 315-332.
  • 5Bertsekas D P. Nonlinear Programming [M]. 2 Ed. Belmont : Athena Scientific, 1999.
  • 6More J J, Garbow B S, Hillstrome K E. Testing Unconstrained Optimization Software [J]. ACM Tran Math Software, 1981,7: 17-41.
  • 7R.Byrd and J.Nocedal. A Tool for the Analysis of Quasi-Newton Methods with Application to Unconstrained Minimization[J]. SIAM Journal on Numerical Analysis,1989(26).
  • 8J.E.Dennes and J.J.More.A Characterization of a Superlinear Convergence and its Application to Quasi-Newton Methods [J]. Math. Comp,1974(28).
  • 9J.FLACHS.On the Convergence,Invariance,and Related Aspects of a Modification of Huang,s Algorthm [J], Journal of Optimization Theory and Applications,1982(37).
  • 10Wei Z,Yu G.Some Progress in Unconstrained Nonlinear Optimization[C]. Proceedings of the 2003, International Conference on Numerical Optimization and Numerical Linear Algebra(Yuan Yaxinag eds),Science Beijing/New York,2004.

引证文献5

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部