Presentation | 1997/7/24 Unique Represetations of Dynamical Systems Produced by Recurrent Nets Masahiro KIMURA, Ryohei NAKANO, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This paper considers learning a dynamical system by a recurrent neural network (RNN). We propose an affine neural dynamical system (A-NDS) as a dynamical system that an RNN actually produces on the output space to approximate a target dynamical system. We present a unique parametric representation of A-NDSs using RNNs and their affine sections with the aim of constructing effective learning algorithms. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | learning of dynamical systems / recurrent net / affine neural dynamical system / uniqure representation |
Paper # | NC97-26 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 1997/7/24(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) | Unique Represetations of Dynamical Systems Produced by Recurrent Nets |
Sub Title (in English) | |
Keyword(1) | learning of dynamical systems |
Keyword(2) | recurrent net |
Keyword(3) | affine neural dynamical system |
Keyword(4) | uniqure representation |
1st Author's Name | Masahiro KIMURA |
1st Author's Affiliation | NTT Communication Science Laboratories() |
2nd Author's Name | Ryohei NAKANO |
2nd Author's Affiliation | NTT Communication Science Laboratories |
Date | 1997/7/24 |
Paper # | NC97-26 |
Volume (vol) | vol.97 |
Number (no) | 201 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |