Fast Computation of Moore-Penrose Inverse Matrices
| dc.creator | Courrieu, Pierre | |
| dc.date | 2008-04-30 | |
| dc.date.accessioned | 2026-07-07T12:18:32Z | |
| dc.date.available | 2026-07-07T12:18:32Z | |
| dc.description | Many neural learning algorithms require to solve large least square systems in order to obtain synaptic weights. Moore-Penrose inverse matrices allow for solving such systems, even with rank deficiency, and they provide minimum-norm vectors of synaptic weights, which contribute to the regularization of the input-output mapping. It is thus of interest to develop fast and accurate algorithms for computing Moore-Penrose inverse matrices. In this paper, an algorithm based on a full rank Cholesky factorization is proposed. The resulting pseudoinverse matrices are similar to those provided by other algorithms. However the computation time is substantially shorter, particularly for large systems. | |
| dc.description | Number of pages: 5. Typo page 26 line 3: one must read W=G^+F (instead of W=G^+W, which does not make sense!) | |
| dc.identifier | https://arxiv.org/abs/0804.4809 | |
| dc.identifier | http://arxiv.org/abs/0804.4809 | |
| dc.identifier | Neural Information Processing - Letters and Reviews 8, 2 (2005) 25-29 | |
| dc.identifier.uri | http://salesiana.dossiersoluciones.com/handle/123456789/212442 | |
| dc.subject | Neural and Evolutionary Computing | |
| dc.title | Fast Computation of Moore-Penrose Inverse Matrices | |
| dc.type | text |