Presentation	1997/12/12 Chaotic Neural Networks Masahiro Endo,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	We propose layered neural networks (LNN) in which chaotic neural networks (CNN) are incorporated, and learning rules for the networks. The CNN is a network reported by Aihara et al. The proposed network is of incorporated CNN in hidden layer of LNN. Although the proposed learning rule is based on the error, back propagation, our neural networks are characterized by intermittent mobilization of CNN. An advantage of this network is that, it can escape from local minima owing to the chaotic behavior of CNN. For LNNs, there has been a traditional way of escaping from local minima where a noise, given from the outside plays an essential role. However, our neural networks work well without relying on external assistance.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	chaos / chaotic neural network / layerd neural network / error back propagation
Paper #	NC97-55
Date of Issue

Conference Information
Committee	NC
Conference Date	1997/12/12(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)	Chaotic Neural Networks
Sub Title (in English)
Keyword(1)	chaos
Keyword(2)	chaotic neural network
Keyword(3)	layerd neural network
Keyword(4)	error back propagation
1st Author's Name	Masahiro Endo
1st Author's Affiliation	Department of Knowledge-based Information Engineering, Toyohashi University of Technology()
Date	1997/12/12
Paper #	NC97-55
Volume (vol)	vol.97
Number (no)	448
Page	pp.pp.-
#Pages	7
Date of Issue