Title :
Positivity and Stability of the dLV Algorithm for Computing Matrix Singular Values
Author :
Iwasaki, Masashi ; Nakamura, Yoshimasa
Author_Institution :
Dept. of Appl. Math. & Phys., Kyoto Univ.
Abstract :
The discrete Lotka-Volterra algorithm for computing matrix singular values is considered. Positivity and boundedness of a parameter and variables are shown to guarantee a numerical stability of the algorithm. Orthogonal polynomials and integrable systems play a key role
Keywords :
Volterra equations; numerical stability; polynomials; singular value decomposition; dLV algorithm positivity; dLV algorithm stability; discrete Lotka-Volterra algorithm; integrable systems; matrix singular values; numerical stability; orthogonal polynomials; Algorithm design and analysis; Books; Convergence; Eigenvalues and eigenfunctions; Informatics; Mathematics; Numerical stability; Physics computing; Polynomials;
Conference_Titel :
Informatics Research for Development of Knowledge Society Infrastructure, 2007. ICKS 2007. Second International Conference on
Conference_Location :
Kyoto
Print_ISBN :
0-7695-2811-2
DOI :
10.1109/ICKS.2007.22