A HYBIRD ALGORITHM OF TWO KINDS OF TRUST REGION METHODS

摘要：

　　In this paper,we propose an algorithm for nonlinear optimization problem that employs both traditional trust region methods.When solving trust region sub-problem,the traditional trust region algorithm is within the trust region centered at the current iteration point,the two improved trust region algorithms are within the trust region centered at one point located in the direction of the negative gradient.When the norm of the gradient k g is less than or equal to 2 10,we use the traditional trust region method,otherwise use the two improved trust region methods.Numerical results are given to show that the algorithm is efficient for the proposed method.

关键词： Unconstrained optimization Trust region methods Negative gradient Trust region radius Trust region sub-problem

作者: Pei-Pei Zhou Qing-Hua Zhou Yue Yang Yue Wang

作者单位: College of mathematics and Computer Science, Hebei University, Baoding, 071002, HebeiProvince, China

会议类型: 国际会议

会议名称: 11th International Symposium on Operations Research and its Applications(第11届运筹学及其应用国际研讨会)

会议地点: 安徽黄山

会议语种:英文

页码: 144-147

在线出版日期: 2013-08-23（万方平台首次上网日期，不代表论文的发表时间）