Milestones in Matrix Computation: The selected works of Gene H. Golub with commentaries

Portada
Raymond Chan, Chen Greif, Dianne O'Leary
OUP Oxford, 22 de febr. 2007 - 584 pàgines
The text presents and discusses some of the most influential papers in Matrix Computation authored by Gene H. Golub, one of the founding fathers of the field. The collection of 21 papers is divided into five main areas: iterative methods for linear systems, solution of least squares problems, matrix factorizations and applications, orthogonal polynomials and quadrature, and eigenvalue problems. Commentaries for each area are provided by leading experts: Anne Greenbaum, Ake Bjorck, Nicholas Higham, Walter Gautschi, and G. W. (Pete) Stewart. Comments on each paper are also included by the original authors, providing the reader with historical information on how the paper came to be written and under what circumstances the collaboration was undertaken. Including a brief biography and facsimiles of the original papers, this text will be of great interest to students and researchers in numerical analysis and scientific computation.
 

Continguts

Publications of Gene H Golub
13
Major Awards
30
References
43
A generalized conjugate gradient method
68
A generalized conjugate gradient method for the numerical
79
Hermitian and SkewHermitian Splitting Methods
102
Commentary by Ake Bjorck
129
References
141
Numerical methods for computing angles between linear
292
Methods for modifying matrix factorizations with
309
Commentary by Walter Gautschi
345
Calculation of Gauss quadrature rules with
359
Matrices moments and quadrature with Gerard
380
Computation of GaussKronrod Quadrature Rules with
434
EIGENVALUE PROBLEMS
452
References
465

Numerical methods for solving linear least
148
Singular value decomposition and least squares solutions
160
The differentiation of pseudoinverses and nonlinear least
181
Generalized crossvalidation as a method for choosing
202
Commentary by Nicholas Higham
227
The simplex method of linear programming using
257
Illconditioned eigensystems and the computation of
485
The block Lanczos method for computing eigenvalues
528
The numerically stable reconstruction of a Jacobi matrix
546
Index
563
Copyright

Altres edicions - Mostra-ho tot

Frases i termes més freqüents

Informació bibliogràfica