Function Approximation Using Generalized Adalines: Fundamental Multi-state Neural Organizations - Couverture souple

Hsu, Pei-Hsun; Wu, Jiann-Ming

 
9783639225723: Function Approximation Using Generalized Adalines: Fundamental Multi-state Neural Organizations
Couverture souple
ISBN 10 : 3639225724 ISBN 13 : 9783639225723
Editeur : VDM Verlag, 2010

Synopsis

Adaline (adaptive linear element) was proposed by Widrow in 1960s and it has been widely applied to construct neural networks in solving tasks of classification, noise cancellation, system identification and signal prediction. An adaline is composed of a receptive field and a threshold function with bipolar output. In this work, we generalize the bipolar threshold function to multi-state transfer function successfully and prove that adaline and perceptron are special cases of it. The supervised learning process is modeled by a mathematical framework mixed with integer and linear programming and solved by a hybrid of mean field annealing and gradient descent methods according to the criteria of minimizing design cost, maximizing utilization of Gaussian units subject to minimal model size. The numerical simulations show that the learning process is able to generate essential internal representations for the mapping underlying training samples.

Les informations fournies dans la section « Synopsis » peuvent faire référence à une autre édition de ce titre.

Présentation de l'éditeur

Adaline (adaptive linear element) was proposed by Widrow in 1960s and it has been widely applied to construct neural networks in solving tasks of classification, noise cancellation, system identification and signal prediction. An adaline is composed of a receptive field and a threshold function with bipolar output. In this work, we generalize the bipolar threshold function to multi-state transfer function successfully and prove that adaline and perceptron are special cases of it. The supervised learning process is modeled by a mathematical framework mixed with integer and linear programming and solved by a hybrid of mean field annealing and gradient descent methods according to the criteria of minimizing design cost, maximizing utilization of Gaussian units subject to minimal model size. The numerical simulations show that the learning process is able to generate essential internal representations for the mapping underlying training samples.

Les informations fournies dans la section « A propos du livre » peuvent faire référence à une autre édition de ce titre.

Éditeur
VDM Verlag
Date d'édition
2010
Langue
anglais
ISBN 10 
3639225724
ISBN 13 
9783639225723
Reliure
Broché
Nombre de pages
64
Coordonnées du fabricant
non disponible
Personne responsable
non disponible