Presentation | 2007-03-16 Analysis of generalization capability on GA based learning of BNNs Akio TAKAHASHI, Hidehiro NAKANO, Arata MIYAUCHI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | BNN can realize any desired Boolean functions provided a sufficient number of hidden neurons. The BNN can be applied to pattern classification, error correcting codes and so on. As a learning method that can reduce hidden layer neurons, many methods have been proposed. But analysis of generalization capability for BNN by these method is not sufficient so far. In this paper, we analize the generalization capability of the conventional and proposed GA-based learning methods to BNNs. Through numerical results, we consider coding and evaluation methods in the GA-based learning, which have good generalization capability. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | 3 layer BNNs / GA-based learning / generalization capability |
Paper # | NC2006-196 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2007/3/9(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) | Analysis of generalization capability on GA based learning of BNNs |
Sub Title (in English) | |
Keyword(1) | 3 layer BNNs |
Keyword(2) | GA-based learning |
Keyword(3) | generalization capability |
1st Author's Name | Akio TAKAHASHI |
1st Author's Affiliation | Musashi Institute of Technology() |
2nd Author's Name | Hidehiro NAKANO |
2nd Author's Affiliation | Musashi Institute of Technology |
3rd Author's Name | Arata MIYAUCHI |
3rd Author's Affiliation | Musashi Institute of Technology |
Date | 2007-03-16 |
Paper # | NC2006-196 |
Volume (vol) | vol.106 |
Number (no) | 590 |
Page | pp.pp.- |
#Pages | 5 |
Date of Issue |