首页 | 本学科首页   官方微博 | 高级检索  
     检索      

大规模半定规划问题的正则化方法及收敛性
引用本文:赵花丽,杨雪梅.大规模半定规划问题的正则化方法及收敛性[J].商洛学院学报,2009,23(2).
作者姓名:赵花丽  杨雪梅
作者单位:咸阳师范学院数学系,陕西咸阳,712000
基金项目:咸阳师范学院科研基金 
摘    要:为了改进求解大规模约束条件的半定规划问题的方法.首先通过经典的二次正则法,将一般的半定规划问题(SDP)的标准形式进行形式的转化.然后通过研究转化后问题的最优性条件,给出了求解一般的半定规划问题的正则化算法及其收敛性证明.在实际中,处理大规模约束条件的半定规划问题(SDP)时,该方法表现出很好的性能.

关 键 词:正则化方法  大规模半定规划问题  收敛性

A Regularization Method and Its Convergence of Large-scale Semidefinite Programming
ZHA Hua-li,YANG Xue-mei.A Regularization Method and Its Convergence of Large-scale Semidefinite Programming[J].Journal of Shangluo University,2009,23(2).
Authors:ZHA Hua-li  YANG Xue-mei
Institution:ZHAO Hua-li,YANG Xue-mei(Department of Mathematics,Xianyang Normal University,Xianyang,Shaanxi 712000)
Abstract:A regularization method introduced by a classical quadratic regularizations is proposed to transform objective function into another SDP.then we study its optimality conditions for solving semidefinite programming(SDP).In the end,a convergence analysis of the algorithm and numerical experiments are given.is studied In practice it behaves very well on some instances of SDP having a large number of constraints.
Keywords:regularization method  large-scale semidefinite programming  convergence  
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号