| Presentation | 2009-03-13 Self-organizing algorithm of dynamical systems : Approaches from Parametric Bias method and mnSOM Takashi OHKUBO, Jun-ichi TSUJI, Kazuhiro TOKUNAGA, Tetsuo FURUKAWA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | The purpose of this work is to develop a neural network architecture which enables us to deal with a group of dynamical systems. The architecture, which is refered as Multi-Dynamics Learning Network (MDLN), is expected to satisfy the following three requirements. (i) By giving a set of time seriese as a training dataset, MDLN estimates those dynamics as well as unobservable variables. (ii) MDLN orders those dynamics according to the similarities between them. (iii) MDLN represents and/or interpolates intermediate dynamics between given dynamics. Though the Recurrent Neural Network with Parametric Bias (RNNPB) and the modular network SOM with recurrent neural network modules (RNN-mnSOM) have been proposed for this purpose, it is shown in this paper that these two architectures do not satisfy the above requirements. It is also shown that an additive normalization algorithm improves the learning performance of RNNPB and RNN-mnSOM so that they satisfy the above three requreiments. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Self-Organizing Map / Recurrent Neural Network / RNNPB / RNN-mnSOM / Dynamics |
| Paper # | NC2008-168 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 2009/3/4(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) | Self-organizing algorithm of dynamical systems : Approaches from Parametric Bias method and mnSOM |
| Sub Title (in English) | |
| Keyword(1) | Self-Organizing Map |
| Keyword(2) | Recurrent Neural Network |
| Keyword(3) | RNNPB |
| Keyword(4) | RNN-mnSOM |
| Keyword(5) | Dynamics |
| 1st Author's Name | Takashi OHKUBO |
| 1st Author's Affiliation | Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology() |
| 2nd Author's Name | Jun-ichi TSUJI |
| 2nd Author's Affiliation | Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology |
| 3rd Author's Name | Kazuhiro TOKUNAGA |
| 3rd Author's Affiliation | Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology |
| 4th Author's Name | Tetsuo FURUKAWA |
| 4th Author's Affiliation | Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology |
| Date | 2009-03-13 |
| Paper # | NC2008-168 |
| Volume (vol) | vol.108 |
| Number (no) | 480 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |