商品簡介
Organizes the many available methods for the numerical solution of eigenvalue problems. The first of the eleven chapters provide the top level of a decision tree for classifying eigenvalue problems, and summarize the mathematical principals of projection onto subspaces and spectral transformations. The next chapters give details for six categories of eigenvalue problems Hermitian, generalized Hermitian, singular value decomposition, non-Hermitian, generalized non-Hermitian, and nonlinear along with algorithm templates and pointers to available software. Common issues of sparse matrix representation and computation shared by all algorithms, and preconditioning techniques complete the volume. Annotation c. Book News, Inc., Portland, OR (booknews.com)
