A limited memory steepest descent method

被引:54
作者
Fletcher, Roger [1 ]
机构
[1] Univ Dundee, Dept Math, Dundee DD1 4HN, Scotland
关键词
QUASI-NEWTON MATRICES; BARZILAI;
D O I
10.1007/s10107-011-0479-6
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
The possibilities inherent in steepest descent methods have been considerably amplified by the introduction of the Barzilai-Borwein choice of step-size, and other related ideas. These methods have proved to be competitive with conjugate gradient methods for the minimization of large dimension unconstrained minimization problems. This paper suggests a method which is able to take advantage of the availability of a few additional 'long' vectors of storage to achieve a significant improvement in performance, both for quadratic and non-quadratic objective functions. It makes use of certain Ritz values related to the Lanczos process (Lanczos in J Res Nat Bur Stand 45:255-282, 1950). Some underlying theory is provided, and numerical evidence is set out showing that the new method provides a competitive and more simple alternative to the state of the art l-BFGS limited memory method.
引用
收藏
页码:413 / 436
页数:24
相关论文
共 26 条
[1]  
[Anonymous], 1987, Unconstrained Optimization: Practical Methods of Optimization
[2]   2-POINT STEP SIZE GRADIENT METHODS [J].
BARZILAI, J ;
BORWEIN, JM .
IMA JOURNAL OF NUMERICAL ANALYSIS, 1988, 8 (01) :141-148
[3]   REPRESENTATIONS OF QUASI-NEWTON MATRICES AND THEIR USE IN LIMITED MEMORY METHODS [J].
BYRD, RH ;
NOCEDAL, J ;
SCHNABEL, RB .
MATHEMATICAL PROGRAMMING, 1994, 63 (02) :129-156
[4]  
Cauchy A-L., 1847, Comp. Rend. Sci. Paris, V25, P536
[5]   On the asymptotic behaviour of some new gradient methods [J].
Dai, YH ;
Fletcher, R .
MATHEMATICAL PROGRAMMING, 2005, 103 (03) :541-559
[6]   Projected Barzilai-Borwein methods for large-scale box-constrained quadratic programming [J].
Dai, YH ;
Fletcher, R .
NUMERISCHE MATHEMATIK, 2005, 100 (01) :21-47
[7]   Analysis of monotone gradient methods [J].
Dai, Yu-Hong ;
Yuan, Ya-xiang .
JOURNAL OF INDUSTRIAL AND MANAGEMENT OPTIMIZATION, 2005, 1 (02) :181-192
[8]  
Fletcher R, 2005, APPL OPTIM, V96, P235
[9]   A RAPIDLY CONVERGENT DESCENT METHOD FOR MINIMIZATION [J].
FLETCHER, R ;
POWELL, MJD .
COMPUTER JOURNAL, 1963, 6 (02) :163-&
[10]  
Fletcher R., 1990, Lect. Appl. Math. (AMS), V26, P165