YANG Qian,WU Zezhong,HE Shengyu.A Kind of Improved BFGS Quasi-Newton Method and its Comparison with other Quasi-Newton Methods[J].Journal of Chengdu University of Information Technology,2023,38(02):227-235.[doi:10.16836/j.cnki.jcuit.2023.02.015]
一类改进的BFGS拟牛顿法及与其他几种拟牛顿法的比较研究
- Title:
- A Kind of Improved BFGS Quasi-Newton Method and its Comparison with other Quasi-Newton Methods
- 文章编号:
- 2096-1618(2023)02-0227-09
- 关键词:
- 无约束最优化; BFGS拟牛顿法; 拟牛顿方程; Wolfe线搜索准则
- Keywords:
- unconstrained optimization; BFGS quasi-Newton method; quasi-Newton equation; Wolfe line search criterion
- 分类号:
- O221.2
- 文献标志码:
- A
- 摘要:
- 拟牛顿法是最优化中一种重要的求解无约束问题的方法,对拟牛顿算法的改进是一个重点研究内容。基于新拟牛顿方程Bk+1•s(k)=(y(k))*提出一个新的改进的BFGS算法,结合以往几种改进的BFGS算法,采用Wolfe线搜索准则对迭代步长进行搜索,并通过选取一些测试函数利用MATLAB工具对这几种改进的BFGS拟牛顿法的收敛效果进行对比分析。实验结果表明,提出的改进的BFGS算法相较于以往几种改进的BFGS算法,收敛效果更好。
- Abstract:
- Quasi-Newton method is an important method to solve unconstrained problems in optimization, and the improvement of the quasi-Newton algorithm is a key research content. In this paper, a new improved BFGS algorithm is proposed based on the new quasi-Newton equationBk+1•s(k)=(y(k))*.Combined with several previous improved BFGS algorithms, the iterative step size is searched by using the Wolfe line search criterion. By selecting some test functions, the convergence effects of these improved BFGS quasi Newton methods are compared and analyzed by using MATLAB tools. The experimental results show that the improved BFGS algorithm proposed in this paper is an improvement with good convergence effect compared with several previous improved BFGS algorithms.
参考文献/References:
[1] Mokhtar S.Bazaraa,Hanif D.Sherali,C.M.Shetty.Nonlinear programming: theory and algorithms[M].Library of Congress Publication,2005.
[2] Wei Z,Li G,Qi L.New quasi-Newton methods for unconstrained optimization problems[J].Computational Optimization and Applications,2006,175:1156-1188.
[3] Wei Z,Yu G,Lian Z.The superlinear convergence of a modified BFGS-type method for unconstrained optimization[J].Computational Optimization and Applications,2004,29:315-332.
[4] Yuan G,Wei Z.Convergence analysis of a modified BFGS method on convex minimizations[J].Computational Optimization and Applications,2010,47:237-255.
[5] 陈奎林.一种改进的BFGS算法及其收敛性分析[J].重庆理工大学学报,2011,25(11):111-114.
[6] Forsgren A,Odland T.On exact linesearch quasi-Newton methods for minimizing a quadratic function[J].Computational Optimization and Applications,2018,69:225-241.
[7] Ek D,Forsgren A.Exact linesearch limited-memory quasi-Newton methods for minimizing a quadratic function[J].Computational Optimization and Applications,2021,79:789-816.
[8] 景慧丽.无约束最优化问题中修改的BFGS方法[J].科技信息,2008,26:209-210.
[9] 袁亚湘,孙文瑜.最优化理论与方法[M].北京:科学出版社,1997.
[10] Yuan G,Wei Z,Lu X.Global convergence of BFGS and PRP methods under a modified weak Wolfe-Powell line search[J].Appl.Math.Modelling,2017,47:811-825.
[11] LI D,FU M.KUSHIMA.A modified BFGS method and its global convergence in nonconvex minimization[J].Journal of Computational and Applied Mathematics,2001,129:15-35.
[12] Momin Jamil,Yang Xin She.A Literature Survey of Benchmark Functions For Global Optimization Problems[J].International Journal of Mathematical Modelling & Numerical Optimisation,2013,4(2):150-194.
相似文献/References:
[1]李菊雯,吴泽忠.基于Armijo搜索步长的BFGS与DFP拟牛顿法的比较研究[J].成都信息工程大学学报,2021,36(05):558.[doi:10.16836/j.cnki.jcuit.2021.05.014]
LI Juwen,WU Zezhong.Comparison of BFGS and DFP Quasi-Newton Method based on Armijo Search Step[J].Journal of Chengdu University of Information Technology,2021,36(02):558.[doi:10.16836/j.cnki.jcuit.2021.05.014]
备注/Memo
收稿日期:2021-12-12
基金项目:国家自然科学基金资助项目(71962030); 四川省社科重点研究基地资助项目(Xq21B06)