| Presentation | 1994/3/25 A realization method of otimally generalizing neural network based on error minimization Hidemitsu Ogawa, Jun-ichi Funada, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | The task required of a Multilayer Perceptron can be summarised as an optimal approximation of the original function from the set of sampled data,incorporating the concept of generalization ability.Based on previous research,it has been shown that even if the original function is not known,it′s best approximation can be computed using the technique namely Projection Generalizing Neural Network(PGNN).However,using the above technique,we obtain a family of NNs realizing the same goal.In this paper,an effort has been made to exploit this existing degree of freedom.While maintaining the optimal generalizing abilty of the net,a method of generating a NN which is least susceptible to errors in the connection weights is theorized. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | multilayer perceptron / generalization / projection learning / projection generalizing heural network / PGNN / minimum error realization |
| Paper # | NC93-127 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 1994/3/25(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) | A realization method of otimally generalizing neural network based on error minimization |
| Sub Title (in English) | |
| Keyword(1) | multilayer perceptron |
| Keyword(2) | generalization |
| Keyword(3) | projection learning |
| Keyword(4) | projection generalizing heural network |
| Keyword(5) | PGNN |
| Keyword(6) | minimum error realization |
| 1st Author's Name | Hidemitsu Ogawa |
| 1st Author's Affiliation | Department of Computer Science,Faculty of Engineering,Tokyo Institute of Technology() |
| 2nd Author's Name | Jun-ichi Funada |
| 2nd Author's Affiliation | Department of Computer Science,Faculty of Engineering,Tokyo Institute of Technology |
| Date | 1994/3/25 |
| Paper # | NC93-127 |
| Volume (vol) | vol.93 |
| Number (no) | 537 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |