Convergence estimates for the generalized Davidson method for symmetric eigenvalue problems I: The preconditioning aspect

被引:12
作者
Ovtchinnikov, E [1 ]
机构
[1] Univ Westminster, Harrow Sch Comp Sci, London HA1 3TP, England
关键词
iterative methods for eigenvalue problems; preconditioning; generalized Davidson method; convergence estimates; superlinear convergence;
D O I
10.1137/S0036142902411756
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
The generalized Davidson (GD) method can be viewed as a generalization of the preconditioned steepest descent (PSD) method for solving symmetric eigenvalue problems. There are two aspects of this generalization. The most obvious one is that in the GD method the new approximation is sought in a larger subspace, namely the one that spans all the previous approximate eigenvectors, in addition to the current one and the preconditioned residual thereof. Another aspect relates to the preconditioning. Most of the available results for the PSD method are associated with the same view on preconditioning as in the case of linear systems. Consequently, they fail to detect the superlinear convergence for certain "ideal" preconditioners, such as the one corresponding to the "exact" version of the Jacobi-Davidson method-one of the most familiar instances of the GD method. Focusing on the preconditioning aspect, this paper advocates an alternative approach to measuring the quality of preconditioning for eigenvalue problems and presents corresponding nonasymptotic convergence estimates for the GD method in general and Jacobi-Davidson method in particular that correctly detect known cases of the superlinear convergence.
引用
收藏
页码:258 / 271
页数:14
相关论文
共 30 条
[1]  
BAI Z., 2000, SOFTWARE ENV TOOLS, V11
[2]   A subspace preconditioning algorithm for eigenvector/eigenvalue computation [J].
Bramble, JH ;
Pasciak, JE ;
Knyazev, AV .
ADVANCES IN COMPUTATIONAL MATHEMATICS, 1996, 6 (02) :159-189
[3]   THE DAVIDSON METHOD [J].
CROUZEIX, M ;
PHILIPPE, B ;
SADKANE, M .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 1994, 15 (01) :62-76
[4]  
D'YAKONOV E. G., 1996, OPTIMIZATION SOLVING
[5]   ITERATIVE CALCULATION OF A FEW OF LOWEST EIGENVALUES AND CORRESPONDING EIGENVECTORS OF LARGE REAL-SYMMETRIC MATRICES [J].
DAVIDSON, ER .
JOURNAL OF COMPUTATIONAL PHYSICS, 1975, 17 (01) :87-94
[6]   ITERATION METHODS IN EIGENVALUE PROBLEMS [J].
DYAKONOV, EG .
MATHEMATICAL NOTES, 1983, 34 (5-6) :945-953
[7]   MINIMIZATION OF THE COMPUTATIONAL LABOR IN DETERMINING THE 1ST EIGENVALUES OF DIFFERENTIAL-OPERATORS [J].
DYAKONOV, EG ;
OREKHOV, MY .
MATHEMATICAL NOTES, 1980, 27 (5-6) :382-391
[8]   Jacobi-Davidson style QR and QZ algorithms for the reduction of matrix pencils [J].
Fokkema, DR ;
Sleijpen, GLG ;
Van der Vorst, HA .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 1998, 20 (01) :94-125
[9]  
Golub GH., 1961, Numer Math, V3, P147, DOI DOI 10.1007/BF01386013
[10]  
Knyazev A. V., 1998, ELECTRON T NUMER ANA, V7, P104