Presentation 2002/3/12
Additional Learnable Neocognitron
Katsuyoshi YANAGAWA, Kunihiko FUKUSHIMA, Toshinobu YOSHIDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We often want to extend pattern recognition neural network. In this paper, we call the method for learning some new pattern to extend already made neural network "additional learning". Because of introduction of new learning rule to neocognitron, this model is able to additional learn. And this model is able to learn while it recognize the pattern. We call this method for learning "serial learning". In this paper, we propose such additional learnable model, and prove this model can (1)serial learn, (2)additional learn unknown category.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) additional learning / serial learning / neocognitron / learning rule / pattern recognition / handwriting recognition
Paper # NC2001-176
Date of Issue

Conference Information
Committee NC
Conference Date 2002/3/12(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Additional Learnable Neocognitron
Sub Title (in English)
Keyword(1) additional learning
Keyword(2) serial learning
Keyword(3) neocognitron
Keyword(4) learning rule
Keyword(5) pattern recognition
Keyword(6) handwriting recognition
1st Author's Name Katsuyoshi YANAGAWA
1st Author's Affiliation Faculty of Electro-Communications, The University of Electro-Communications()
2nd Author's Name Kunihiko FUKUSHIMA
2nd Author's Affiliation Katayanagi Advanced Research Laboratories, Tokyo University of Technology
3rd Author's Name Toshinobu YOSHIDA
3rd Author's Affiliation Faculty of Electro-Communications, The University of Electro-Communications
Date 2002/3/12
Paper # NC2001-176
Volume (vol) vol.101
Number (no) 736
Page pp.pp.-
#Pages 8
Date of Issue