Pattern Association: A Reference Book: A hybrid approach for pattern association - Couverture souple

Kumar, Somesh

 
9783330021198: Pattern Association: A Reference Book: A hybrid approach for pattern association
Couverture souple
ISBN 10 : 3330021195 ISBN 13 : 9783330021198
Editeur : LAP LAMBERT Academic Publishing, 2017

Synopsis

In the present book, our objective is to develop a hybrid evolutionary system consisting of Hopfield neural network and genetic algorithm which will be responsible for evolution of weight matrices in order to store some input patterns and analyze the performance of such a system in the terms of correct recalling of these already stored patterns again with evolutionary algorithm by presenting the same or noisy versions of input patterns. In this process, first the patterns of training set have been encoded in the neural network using MC-adaptation rule. It is expected that all the patterns of training set has been successfully stored as the associative memory feature of Hopfield type neural network. As a result of this learning process, we obtain the expected optimized weight matrices. Now, we employ the genetic algorithm to evolve the population of these approximate optimal weight matrices obtained by MC-adaptation rule. The fitness of every evolved population of weight matrices is evaluated by using the two fitness evaluation functions.

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

Présentation de l'éditeur

In the present book, our objective is to develop a hybrid evolutionary system consisting of Hopfield neural network and genetic algorithm which will be responsible for evolution of weight matrices in order to store some input patterns and analyze the performance of such a system in the terms of correct recalling of these already stored patterns again with evolutionary algorithm by presenting the same or noisy versions of input patterns. In this process, first the patterns of training set have been encoded in the neural network using MC-adaptation rule. It is expected that all the patterns of training set has been successfully stored as the associative memory feature of Hopfield type neural network. As a result of this learning process, we obtain the expected optimized weight matrices. Now, we employ the genetic algorithm to evolve the population of these approximate optimal weight matrices obtained by MC-adaptation rule. The fitness of every evolved population of weight matrices is evaluated by using the two fitness evaluation functions.

Biographie de l'auteur

Dr. Somesh Kumar presently spearheads the MCA as a Director and IT department in NIET, Greater Noida, India. He has completed his MCA degree in 2000, ME (CS&E) in 2006, and Ph.D(CS) in 2011. Since 2000, he is in teaching profession and published a no. of research papers in journals of repute like Applied Soft Computing, Connection Science, etc.

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

Éditeur
LAP LAMBERT Academic Publishing
Date d'édition
2017
Langue
anglais
ISBN 10 
3330021195
ISBN 13 
9783330021198
Reliure
Broché
Nombre de pages
108
Coordonnées du fabricant
non disponible
Personne responsable
KreGa Onlinehandel GmbH
4950000000000
info@krega-onlinehandel.de

Birkenweg 31
Brakel
33034
Allemagne