In this paper, we present a new class of quasi-Newton methods for the effective learning in large multilayer perceptron (MLP)-networks. The algorithms introduced in this work, named LQN, utilize an iterative scheme of a generalized BFGS-type method, involving a suitable family of matrix algebras L. The main advantages of these innovative methods are based upon the fact that they have an O(n log_2 n) complexity per step and that they require O(n) memory allocations. Numerical experiences, performed on a set of standard benchmarks of MLP-networks, show the competitivity of the LQN methods, especially for large values of n.
Bortoletti, A., Di Fiore, C., Fanelli, S., & Zellini, P. (2003). A new class of quasi-newtonian methods for optimal learning in MLP-networks. IEEE TRANSACTIONS ON NEURAL NETWORKS, 14(2), 263-273.
Tipologia: | Articolo su rivista |
Citazione: | Bortoletti, A., Di Fiore, C., Fanelli, S., & Zellini, P. (2003). A new class of quasi-newtonian methods for optimal learning in MLP-networks. IEEE TRANSACTIONS ON NEURAL NETWORKS, 14(2), 263-273. |
IF: | Con Impact Factor ISI |
Lingua: | English |
Settore Scientifico Disciplinare: | Settore MAT/08 - Analisi Numerica |
Revisione (peer review): | Sì, ma tipo non specificato |
Tipo: | Articolo |
Rilevanza: | Rilevanza internazionale |
Stato di pubblicazione: | Pubblicato |
Data di pubblicazione: | 2003 |
Titolo: | A new class of quasi-newtonian methods for optimal learning in MLP-networks |
Autori: | |
Autori: | Bortoletti, A; Di Fiore, C; Fanelli, S; Zellini, P |
Appare nelle tipologie: | 01 - Articolo su rivista |