Result: Neocognitron trained by winner-kill-loser with triple threshold
Title:
Neocognitron trained by winner-kill-loser with triple threshold
Authors:
Source:
Neurocomputing (Amsterdam). 129:78-84
Publisher Information:
Amsterdam: Elsevier, 2014.
Publication Year:
2014
Physical Description:
print, 7 ref
Original Material:
INIST-CNRS
Subject Terms:
Cognition, Computer science, Informatique, Sciences exactes et technologie, Exact sciences and technology, Sciences appliquees, Applied sciences, Informatique; automatique theorique; systemes, Computer science; control theory; systems, Intelligence artificielle, Artificial intelligence, Reconnaissance des formes. Traitement numérique des images. Géométrie algorithmique, Pattern recognition. Digital image processing. Computational geometry, Connexionnisme. Réseaux neuronaux, Connectionism. Neural networks, Sciences biologiques et medicales, Biological and medical sciences, Sciences biologiques fondamentales et appliquees. Psychologie, Fundamental and applied biological sciences. Psychology, Psychologie. Psychophysiologie, Psychology. Psychophysiology, Langage, Language, Psychologie. Psychanalyse. Psychiatrie, Psychology. Psychoanalysis. Psychiatry, Intelligence artificielle, Artificial intelligence, Inteligencia artificial, Reconnaissance caractère, Character recognition, Reconocimiento carácter, Reconnaissance forme, Pattern recognition, Reconocimiento patrón, Reconnaissance image, Image recognition, Reconocimiento imagen, Réseau multicouche, Multilayer network, Red multinivel, Réseau neuronal, Neural network, Red neuronal, Simulation ordinateur, Computer simulation, Simulación computadora, Système hiérarchisé, Hierarchical system, Sistema jerarquizado, Tout le gain au vainqueur, Winner take all, Toda ganancia al vencedor, Traitement image, Image processing, Procesamiento imagen, Vision ordinateur, Computer vision, Visión ordenador, Algorithme compétitif, Competitive algorithms, Algoritmo Competitivo, Hierarchical network, Neocognitron, Triple threshold, Visual pattern recognition, Winner-kill-loser
Document Type:
Conference
Conference Paper
File Description:
text
Language:
English
Author Affiliations:
Fuzzy Logic Systems Institute, Iizuka, Fukuoka 820-0067, Japan
Faculty of Informatics, Kansai University, Takatsuki, Osaka 569-1095, Japan
Center for Computational Neuroscience and Neural Technology, Boston University, Boston, MA 02215, United States
Faculty of Informatics, Kansai University, Takatsuki, Osaka 569-1095, Japan
Center for Computational Neuroscience and Neural Technology, Boston University, Boston, MA 02215, United States
ISSN:
0925-2312
Rights:
Copyright 2015 INIST-CNRS
CC BY 4.0
Sauf mention contraire ci-dessus, le contenu de cette notice bibliographique peut être utilisé dans le cadre d’une licence CC BY 4.0 Inist-CNRS / Unless otherwise stated above, the content of this bibliographic record may be used under a CC BY 4.0 licence by Inist-CNRS / A menos que se haya señalado antes, el contenido de este registro bibliográfico puede ser utilizado al amparo de una licencia CC BY 4.0 Inist-CNRS
CC BY 4.0
Sauf mention contraire ci-dessus, le contenu de cette notice bibliographique peut être utilisé dans le cadre d’une licence CC BY 4.0 Inist-CNRS / Unless otherwise stated above, the content of this bibliographic record may be used under a CC BY 4.0 licence by Inist-CNRS / A menos que se haya señalado antes, el contenido de este registro bibliográfico puede ser utilizado al amparo de una licencia CC BY 4.0 Inist-CNRS
Notes:
Computer science; theoretical automation; systems
Psychology. Ethology
FRANCIS
Psychology. Ethology
FRANCIS
Accession Number:
edscal.28284370
Database:
PASCAL Archive
Further Information
The neocognitron is a hierarchical, multi-layered neural network capable of robust visual pattern recognition. The neocognitron acquires the ability to recognize visual patterns through learning. The winner-kill-loser is a competitive learning rule recently shown to outperform standard winner-take-all learning when used in the neocognitron to perform a character recognition task. In this paper, we improve over the winner-kill-loser rule by introducing an additional threshold to the already existing two thresholds used in the original version. It is shown theoretically, and also by computer simulation, that the use of a triple threshold makes the learning process more stable. In particular, a high recognition rate can be obtained with a smaller network.