Presentation 1998/3/20
An Efficient Way of Using Layered Neural Networks for Multiclass Problems
Seiji Ishihara, Takashi Nagano,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In multiclass problems, the increase of class number causes increase of iteration and decrease of recognition rate when utilizing a feedforward neural network with back-propagation algorithm. Furthermore the total number of parameters in the network increases. Our approach to avoid these troubles is to use a modular network architecture. The modular network consists of small modules. In this paper we present that the total number of parameters in the modular network proposed by us becomes smaller than that in the usual non-modular network as the number of classes increases.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) layered neural network / modular network / multiclass problem
Paper #
Date of Issue

Conference Information
Committee NC
Conference Date 1998/3/20(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) An Efficient Way of Using Layered Neural Networks for Multiclass Problems
Sub Title (in English)
Keyword(1) layered neural network
Keyword(2) modular network
Keyword(3) multiclass problem
1st Author's Name Seiji Ishihara
1st Author's Affiliation Department of Industrial and System Engineering, College of Engineering, Hosei University()
2nd Author's Name Takashi Nagano
2nd Author's Affiliation Department of Industrial and System Engineering, College of Engineering, Hosei University
Date 1998/3/20
Paper #
Volume (vol) vol.97
Number (no) 624
Page pp.pp.-
#Pages 6
Date of Issue