| Presentation | 1996/12/14 The least squared error of layered neural networks for a Gaussian noise sequence Katsuyuki HAGIWARA, Kazuhiro KUNO, Shiro USUI, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In the model selection problem of layered neural networks, it is known that there exists a case in which the relationship between learning error and generalization error can not be derived by means of the asymptotic expansion due to the degeneration of the Fisher information matrlx, which is caused by nonuniqueness of the connection weights. The typical data in such case is a Gaussian noise sequence. For such data, by focusing on the adaptability of basis functions in layered neural networks, we have been analyzed the least square error and the predictive square error of the model with a finite set of basis functions and some additional imposed conditions. In this paper, we derive a upper bound on the expectation of the least square error with respect to the distribution of the data for some types of networks by applying the asymptotic results obtained previously. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | adaptive basis function / the least squared error / Radial Basis Function / bell type output function / sigmoid type output function |
| Paper # | NC96-59 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 1996/12/14(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) | The least squared error of layered neural networks for a Gaussian noise sequence |
| Sub Title (in English) | |
| Keyword(1) | adaptive basis function |
| Keyword(2) | the least squared error |
| Keyword(3) | Radial Basis Function |
| Keyword(4) | bell type output function |
| Keyword(5) | sigmoid type output function |
| 1st Author's Name | Katsuyuki HAGIWARA |
| 1st Author's Affiliation | Faculty of Electrical & Electronic Engineering, Mie University() |
| 2nd Author's Name | Kazuhiro KUNO |
| 2nd Author's Affiliation | Faculty of Electrical & Electronic Engineering, Mie University |
| 3rd Author's Name | Shiro USUI |
| 3rd Author's Affiliation | Department of Information and Computer Sciences, Toyohashi University of Technology |
| Date | 1996/12/14 |
| Paper # | NC96-59 |
| Volume (vol) | vol.96 |
| Number (no) | 430 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |