Presentation | 2003/11/14 An Evaluation Method of Generalization Function of Artificial Neural Network Using Riemannian Geometry Ryo MOTOMURA, Hirokazu YOKOI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In the case of multiple conditioning for artificial neural network, generalization characteristics can be judged by describing an output in two-dimensional curved surface, if input of network is to two. Then, effective and new evaluation method using Reimanian geometry for evaluation of generalization function case in which input variable is n piece was proposed in this study. Computer simulation of classical conditioning was carried out on F-LHT network, F-LDT network, and compound network in order to confirm effectiveness of this technique, when input variable was two. As the result, F-LDT network and compound network showed Pavlov conditioning peculiar characteristics. Similar result was got even in evaluation using Riemannian geometric under way quantity, and it was possible to grasp shape of curved surface. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Pavlov conditioning generalization / generalization function / Riemannian geometry |
Paper # | NC2003-72 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2003/11/14(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) | An Evaluation Method of Generalization Function of Artificial Neural Network Using Riemannian Geometry |
Sub Title (in English) | |
Keyword(1) | Pavlov conditioning generalization |
Keyword(2) | generalization function |
Keyword(3) | Riemannian geometry |
1st Author's Name | Ryo MOTOMURA |
1st Author's Affiliation | Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology() |
2nd Author's Name | Hirokazu YOKOI |
2nd Author's Affiliation | Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology |
Date | 2003/11/14 |
Paper # | NC2003-72 |
Volume (vol) | vol.103 |
Number (no) | 465 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |