Superlinearly convergent trust-region method without the assumption of positive-definite Hessian
Zhang, J. L.; Wang, Y.; Zhang, X. S.
刊名JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS
2006-04-01
卷号129期号:1页码:201-218
关键词trust-region methods trust-region radius global convergence superlinear convergence local error bound
ISSN号0022-3239
DOI10.1007/s10957-006-9053-4
英文摘要In this paper, we reinvestigate the trust-region method by reformulating its subproblem: the trust-region radius is guided by gradient information at the current iteration and is self-adaptively adjusted. A trust-region algorithm based on the proposed subproblem is proved to be globally convergent. Moreover, the superlinear convergence of the new algorithm is shown without the condition that the Hessian of the objective function at the solution be positive definite. Preliminary numerical results indicate that the performance of the new method is notable.
WOS研究方向Operations Research & Management Science ; Mathematics
语种英语
出版者SPRINGER/PLENUM PUBLISHERS
WOS记录号WOS:000242660000012
内容类型期刊论文
源URL[http://ir.amss.ac.cn/handle/2S8OKBNM/3443]  
专题中国科学院数学与系统科学研究院
通讯作者Zhang, J. L.
作者单位1.Grad Univ, Sch Management, Chinese Acad Sci, Res Ctr Data Tech & Knowledge Econ, Beijing, Peoples R China
2.Chinese Acad Sci, Acad Math & Syst Sci, Inst Appl Math, Beijing, Peoples R China
推荐引用方式
GB/T 7714
Zhang, J. L.,Wang, Y.,Zhang, X. S.. Superlinearly convergent trust-region method without the assumption of positive-definite Hessian[J]. JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS,2006,129(1):201-218.
APA Zhang, J. L.,Wang, Y.,&Zhang, X. S..(2006).Superlinearly convergent trust-region method without the assumption of positive-definite Hessian.JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS,129(1),201-218.
MLA Zhang, J. L.,et al."Superlinearly convergent trust-region method without the assumption of positive-definite Hessian".JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS 129.1(2006):201-218.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace