Presentation | 2013-01-29 Multi-layered GMDH-type neural network algorithm using Prediction Sum ofSquares(PSS)criterion and its applicationt ononlinear system identification Tadashi Kondo, Junji Ueno, Shoichiro Takao, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this study, a revised Group Method of Data Handling (GMDH)-type neural network using Prediction Sum of Squares (PSS) criterion is proposed and applied to the identification problem of the nonlinear system Revised GMDH-type neural network algorithm can organize the neural network architectures using the heuristic self-organization method which is a kind of evolutional computation It has an ability of self-selecting optimum neural network architecture from three neural network architectures such as sigmoid function neural network, radial basis function (RBF) neural network and polynomial neural network and also has abilities of self-selecting the number of layers, the number of neurons in hidden layers and useful input variables so as to minimize PSS criterion Furthermore, in this algorithm, the principal component-regression analysis is used to learning calculation of the weights and there is no multi co-linearity in the neural network architectures This algorithm is applied to the nonlinear system identification problem and it is shown that this algorithm is accurate and useful for the nonlinear system identification. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | GMDH / Neural network / Nonlinear system identification |
Paper # | KBSE2012-64 |
Date of Issue |
Conference Information | |
Committee | KBSE |
---|---|
Conference Date | 2013/1/21(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 | Knowledge-Based Software Engineering (KBSE) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Multi-layered GMDH-type neural network algorithm using Prediction Sum ofSquares(PSS)criterion and its applicationt ononlinear system identification |
Sub Title (in English) | |
Keyword(1) | GMDH |
Keyword(2) | Neural network |
Keyword(3) | Nonlinear system identification |
1st Author's Name | Tadashi Kondo |
1st Author's Affiliation | Graduate School of Health Sciences, The University of Tokushima() |
2nd Author's Name | Junji Ueno |
2nd Author's Affiliation | Graduate School of Health Sciences, The University of Tokushima |
3rd Author's Name | Shoichiro Takao |
3rd Author's Affiliation | Graduate School of Health Sciences, The University of Tokushima |
Date | 2013-01-29 |
Paper # | KBSE2012-64 |
Volume (vol) | vol.112 |
Number (no) | 419 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |