| Presentation | 1999/3/18 Novel Patterns Learning by Utilizing the Bifurcation of the Internal State of Network Kazuhiro KOJIMA, Koji ITO, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In conventional associative model such as Hopfield model, the information processes have to be divided into the two process, i.e., associative phase changed by the external observers. But there are not any external observers in brain and neural systems. We propose the autonomous novel pattern learning method by utilizing the bifurcation of the internal state of the network. We constructed the neural network from Lorenz systems to realize "I don't know" state. Further, in computer simulatoions, the dynamical properties of the network are studied. It is shown that when no embedded pattern, i.e , unknown pattern, is given to the network, the network state oscillates chaotically. It is considered as "I don't know" state. Further, when the proposed learning method is applied to the network, the network can change the associative phase and learning phase autonomously. Furthermore, the network can learn new patterns without destroying the previous embedded patterns. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Chaotic Dynamics / Mutually Combined Lorenz System / I don't know state / Bifurcation / Autonomous Learning |
| Paper # | NC98-139 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 1999/3/18(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) | Novel Patterns Learning by Utilizing the Bifurcation of the Internal State of Network |
| Sub Title (in English) | |
| Keyword(1) | Chaotic Dynamics |
| Keyword(2) | Mutually Combined Lorenz System |
| Keyword(3) | I don't know state |
| Keyword(4) | Bifurcation |
| Keyword(5) | Autonomous Learning |
| 1st Author's Name | Kazuhiro KOJIMA |
| 1st Author's Affiliation | Interdisciplinary Graduate School of Science and Engineering, Department of Computational Intelligence and Systems Science, Tokyo Institude of Technology() |
| 2nd Author's Name | Koji ITO |
| 2nd Author's Affiliation | Interdisciplinary Graduate School of Science and Engineering, Department of Computational Intelligence and Systems Science, Tokyo Institude of Technology |
| Date | 1999/3/18 |
| Paper # | NC98-139 |
| Volume (vol) | vol.98 |
| Number (no) | 673 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |