Presentation 1994/12/16
Transient analysis of stiff circuits by using modified backward Euler algorithm
Hiroshi Kubo, Hiroyuki Nakajima, Yoshisuke Ueda,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Learning continuous dynamical systems with a scalar-valued teacher signal by threelayered recurrent neural networks(RNN)is argued.The learning of dynamical systems is defined as giving appropriate connection weights so that a RNN and the target system become C^∞ -conjugate.Next,It is pointed out that diffeomorphism f rom the state space of the target system to that of the RNN can be described with operators on scalar functions.Finally,a RNN with a structure on the basis of the convolution operator is proposed and some numerical examples are shown.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) neural network / continuous dyhamical System / operator / Convolution / embedding
Paper # NLP94-72
Date of Issue

Conference Information
Committee NLP
Conference Date 1994/12/16(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 Nonlinear Problems (NLP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Transient analysis of stiff circuits by using modified backward Euler algorithm
Sub Title (in English)
Keyword(1) neural network
Keyword(2) continuous dyhamical System
Keyword(3) operator
Keyword(4) Convolution
Keyword(5) embedding
1st Author's Name Hiroshi Kubo
1st Author's Affiliation Department of Electricul Engineering II,Faculty of Engineering, Kyoto University()
2nd Author's Name Hiroyuki Nakajima
2nd Author's Affiliation Department of Electricul Engineering II,Faculty of Engineering, Kyoto University
3rd Author's Name Yoshisuke Ueda
3rd Author's Affiliation Department of Electricul Engineering II,Faculty of Engineering, Kyoto University
Date 1994/12/16
Paper # NLP94-72
Volume (vol) vol.94
Number (no) 418
Page pp.pp.-
#Pages 8
Date of Issue