Presentation | 1999/6/18 On the Additional Learning by a Multi-Layer Hybrid Neural Network Model Tomoyuki Ogawa, Yasushi Hibino, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Adaptive Resonance Theory (ART) using the competitive learning provides a way to resolve the "stability-plasticity dilemma" that many of neural network models have. The ART, however, has an essential problem that it cannot classify the linearly unseparable patterns. On the other hand, it is known that combining plural models gives new features which are not obtained by a single model. In this paper, a novel multi-layer hybrid neural network model, named Adaptive Category Unifying Network (AC_TUN) is proposed. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | hybrid neural network / perceptron / additional learning / ART / competitive learning |
Paper # | NC99-11 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 1999/6/18(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) | On the Additional Learning by a Multi-Layer Hybrid Neural Network Model |
Sub Title (in English) | |
Keyword(1) | hybrid neural network |
Keyword(2) | perceptron |
Keyword(3) | additional learning |
Keyword(4) | ART |
Keyword(5) | competitive learning |
1st Author's Name | Tomoyuki Ogawa |
1st Author's Affiliation | Computer Technology Integrator Co., Ltd.() |
2nd Author's Name | Yasushi Hibino |
2nd Author's Affiliation | Japan Advanced Institute of Science and Technology |
Date | 1999/6/18 |
Paper # | NC99-11 |
Volume (vol) | vol.99 |
Number (no) | 131 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |