• 论文 •    下一篇

基于修正拟牛顿方程的两阶段步长非单调稀疏对角变尺度梯度投影算法

孙清滢1, 段立宁1, 陈颖梅1, 王宣战1, 宫恩龙2, 徐胜来2   

  1. 1. 中国石油大学(华东),理学院, 山东青岛 266580;
    2. 青岛酒店管理职业技术学院, 山东青岛 266100
  • 收稿日期:2011-06-13 出版日期:2013-05-15 发布日期:2013-05-13
  • 基金资助:

    国家自然科学基金(10971118, 61201455)项目和中央高校基本 科研业务费专项资金(10CX04044A, 11CX06087A).

孙清滢, 段立宁, 陈颖梅, 王宣战, 宫恩龙, 徐胜来. 基于修正拟牛顿方程的两阶段步长非单调稀疏对角变尺度梯度投影算法[J]. 计算数学, 2013, 35(2): 113-124.

Sun Qingying, Duan Lining, Chen Yingmei, Wang Xuanzhan, Gong Enlong, Xu Shenglai. NON-MONOTONE TWO STAGES STEPSIZE DIAGONAL SPARSE VARIABLE METRIC GRADIENT PROJECTION METHOD BASED ON MODIFIED QUASI-NEWTON EQUATION[J]. Mathematica Numerica Sinica, 2013, 35(2): 113-124.

NON-MONOTONE TWO STAGES STEPSIZE DIAGONAL SPARSE VARIABLE METRIC GRADIENT PROJECTION METHOD BASED ON MODIFIED QUASI-NEWTON EQUATION

Sun Qingying1, Duan Lining1, Chen Yingmei1, Wang Xuanzhan1, Gong Enlong2, Xu Shenglai2   

  1. 1. College of Science, China University of Petroleum, Qingdao 266580, Shandong, China;
    2. Qingdao Hotel Management College, Qingdao 266100, Shandong, China
  • Received:2011-06-13 Online:2013-05-15 Published:2013-05-13
基于修正拟牛顿方程, 利用Goldstein-Levitin-Polyak (GLP)投影技术, 建立了 求解带凸集约束的优化问题的两阶段步长Zhang H.C.非单调变尺度梯度投影方法, 证明了算法的全局收敛性. 数值实验表明算法是有效的, 适合求解大规模问题.
Based on modified quasi-Newton equation, by combining with Goldstein- Levitin- Polyak (GLP) projection technique, a new Zhang H.C. non-monotone two stages stepsize diagonal sparse variable metric gradient projection method for nonlinear optimization problem is presented. The global convergence properties of the new method are proved. The numerical results show that the new method is effective and is fit to solve large-scale problems.

MR(2010)主题分类: 

()
[1] Goldstein A A. Convex programming in Hilbert space[J]. Bulletin of the American MathematicalSociety, 1964, 70: 709-710.

[2] Levitin E S, Polyak B T. Constrained minimization problems[J]. USSR Computational Mathematicsand Mathematical Physics, 1966, 6: 1-50.

[3] Calamai P H, Moré J J. Projected gradient methods for linearly constrained problems[J]. Mathmatical.Programming, 1987, 39: 93-116.

[4] Rustem B. Projection methods in constrained optimisation and applications to optimal policydecisions. Springer-Vertag. Berlin, 1981.

[5] Allwright J C. A feasible direction algorithm for convex optimization: global convergence rates[J].Journal of Optimization Theory and Applications, 1980, 1-18.

[6] Psenichny B N, Danilin Y M. Numerical methods in extremal problems. Mir Publishers. Moscow,1978.

[7] Rustem B. A class of superlinearly convergent projection algorithms with relaxed stepsizes[J].Applied Mathematics and Optimization, 1984, 12: 29-43.

[8] Barzilai J and Borwein J M. Two-point step size gradient methods[J]. IMA Journal of NumericalAnalysis, 1988, 8: 141-148.

[9] Raydan M. On the Barzilai and Borwein choice of steplength for the gradient method[J]. IMAJournal of Numerical Analysis, 1993, 13: 321-326.

[10] Raydan M and Svaiter B F. Relaxed steepest descent and Cauchy- Barzilai-Borwein method[J].Computational Optimization and Applications, 2002, 21: 155-167.

[11] Raydan M. The Barzilai and Borwein gradient method for large scale unconstrained minimizationproblems [J], SIAM Journal on Optimization, 1997, 7(1): 26-33.

[12] 时贞军, 孙国. 无约束优化问题的对角稀疏拟牛顿算法[J]. 系统科学与数学, 2006, 26(1): 101-112.

[13] Zengxin Wei, Guoyin Li, Liqun Qi. New quasi-Newton methods for unconstrained optimizationproblems[J]. Applied Mathematics and Computation, 2006, 175: 1156-1188.

[14] Grippo L, Lampariello F and Lucidi S. A nonmonotone line search technique for Newton’smethod[J]. SIAM Journal on Numerical Analysis, 1986, 23(4): 707-716.

[15] Dai Y H. On the nonmonotone line search[J]. Journal of Optimization Theory and Applications,2002, 112: 315-330.

[16] Dai Y H. A nonmonotone conjugate gradient algorithm for unconstrained optimization[J]. Journalof Systems Science and Complexity, 2002, 15: 139-145.

[17] Sun Qingying. Global convergence results of a three term memory gradient method with a nonmonotoneline search technique[J]. ACTA Mathematica Scientia, 2005, 25B(1): 170-178.

[18] 孙清滢, 崔彬, 王长钰. 新非单调线搜索规则的Lampariello修正对角稀疏拟牛顿算法[J].计算数学, 2008, 30(3): 255-268.

[19] 孙清滢, 郑艳梅. 大步长非单调线搜索规则的Lampariello修正对角稀疏拟牛顿算法的全局收敛性[J]. 数学进展, 2008, 37(3): 311-320.

[20] Zhang H C, HagerWW. A nonmonotone line search technique and its application to unconstrainedoptimization[J]. SIAM Journal on Optimization, 2004, 14(4): 1043-1056.

[21] Conn A R, Gould N I M and Toint PH L. Testing a class of methods for solving minimizationproblems with simple bounds on the variable[J]. Mathematics of Computation, 1988, 50: 399-430.
[1] 古振东, 孙丽英. 非线性第二类Volterra积分方程的Chebyshev谱配置法[J]. 计算数学, 2020, 42(4): 445-456.
[2] 尹江华, 简金宝, 江羡珍. 凸约束非光滑方程组一个新的谱梯度投影算法[J]. 计算数学, 2020, 42(4): 457-471.
[3] 张纯, 贾泽慧, 蔡邢菊. 广义鞍点问题的改进的类SOR算法[J]. 计算数学, 2020, 42(1): 39-50.
[4] 彭捷, 代新杰, 肖爱国, 卜玮平. 中立型随机延迟微分方程分裂步θ方法的强收敛性[J]. 计算数学, 2020, 42(1): 18-38.
[5] 李枝枝, 柯艺芬, 储日升, 张怀. 二阶锥线性互补问题的广义模系矩阵分裂迭代算法[J]. 计算数学, 2019, 41(4): 395-405.
[6] 胡冬冬, 曹学年, 蒋慧灵. 带非线性源项的双侧空间分数阶扩散方程的隐式中点方法[J]. 计算数学, 2019, 41(3): 295-307.
[7] 盛秀兰, 赵润苗, 吴宏伟. 二维线性双曲型方程Neumann边值问题的紧交替方向隐格式[J]. 计算数学, 2019, 41(3): 266-294.
[8] 岳超. 高阶分裂步(θ1,θ2,θ3)方法的强收敛性[J]. 计算数学, 2019, 41(2): 126-155.
[9] 杨晋平, 李志强, 闫玉斌. 求解Riesz空间分数阶扩散方程的一种新的数值方法[J]. 计算数学, 2019, 41(2): 170-190.
[10] 王俊俊, 李庆富, 石东洋. 非线性抛物方程混合有限元方法的高精度分析[J]. 计算数学, 2019, 41(2): 191-211.
[11] 张维, 王文强. 随机微分方程改进的分裂步单支θ方法的强收敛性[J]. 计算数学, 2019, 41(1): 12-36.
[12] 王同科, 樊梦. 第二类端点奇异Fredholm积分方程的分数阶退化核方法[J]. 计算数学, 2019, 41(1): 66-81.
[13] 王志强, 文立平, 朱珍民. 时间延迟扩散-波动分数阶微分方程有限差分方法[J]. 计算数学, 2019, 41(1): 82-90.
[14] 陈圣杰, 戴彧虹, 徐凤敏. 稀疏线性规划研究[J]. 计算数学, 2018, 40(4): 339-353.
[15] 姜帆, 刘雅梅, 蔡邢菊. 一类自适应广义交替方向乘子法[J]. 计算数学, 2018, 40(4): 367-386.
阅读次数
全文


摘要