A new complete procedure for the selection of pruning threshold in MIMO (Multiple Input Multiple Output) Feedforward Artificial Neural Networks (FANN) is presented. It is based on the evaluation of a local sensitivity index calculated with respect of any single output of the network. Special emphasis is given to a particular class of neural networks with multiple heterogeneous outputs. It will be shown how to take into account of the non-homogeneous nature of the outputs by deriving an “importance index” from the nonlinear correlation of data. An example of the proposed method will be shown by the development of a neural architecture devoted to a specific multi-output inversion system.
A Pruning Method for Multiple Heterogeneous Output Neural Networks / F. Grasso; A. Luchetta; S. Manetti. - ELETTRONICO. - (2008), pp. 7-2-7-7. (Intervento presentato al convegno IEEE Int. Conference on Intelligent Systems (IS’08) tenutosi a Varna, Bulgaria nel Settembre 2008) [10.1109/IS.2008.4670442].
A Pruning Method for Multiple Heterogeneous Output Neural Networks
GRASSO, FRANCESCO;LUCHETTA, ANTONIO;MANETTI, STEFANO
2008
Abstract
A new complete procedure for the selection of pruning threshold in MIMO (Multiple Input Multiple Output) Feedforward Artificial Neural Networks (FANN) is presented. It is based on the evaluation of a local sensitivity index calculated with respect of any single output of the network. Special emphasis is given to a particular class of neural networks with multiple heterogeneous outputs. It will be shown how to take into account of the non-homogeneous nature of the outputs by deriving an “importance index” from the nonlinear correlation of data. An example of the proposed method will be shown by the development of a neural architecture devoted to a specific multi-output inversion system.File | Dimensione | Formato | |
---|---|---|---|
IS2008_Luchetta.pdf
Accesso chiuso
Tipologia:
Versione finale referata (Postprint, Accepted manuscript)
Licenza:
Tutti i diritti riservati
Dimensione
266.05 kB
Formato
Adobe PDF
|
266.05 kB | Adobe PDF | Richiedi una copia |
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.