| Presentation | 1998/2/5 Approximation Capability of Neural Networks with Time-Delayed Feedbacks Isao Tokuda, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We present an algorithm for constructing continuous-time recurrent neural network with time delayed feedbacks(DRNN) which approximates the dynamics of any differential difference equation of the form x=F(x(t), x(t-γ_1), .., x(t-γ_d)). Efficiency of the algorithm is demonstrated by numerical experiments using the Mackey-Glass equation and the Rossler equation. Based on the constructive algorithm, class of dynamical systems approximated by DRNNs is discussed. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | approximation capability / supervised learning / recurrent neural network with time-delayed feedbacks / functional differential equation |
| Paper # | |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 1998/2/5(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 | ENG |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Approximation Capability of Neural Networks with Time-Delayed Feedbacks |
| Sub Title (in English) | |
| Keyword(1) | approximation capability |
| Keyword(2) | supervised learning |
| Keyword(3) | recurrent neural network with time-delayed feedbacks |
| Keyword(4) | functional differential equation |
| 1st Author's Name | Isao Tokuda |
| 1st Author's Affiliation | Department of Computer Science and Systems Engineering, Muroran Institute of Technology() |
| Date | 1998/2/5 |
| Paper # | |
| Volume (vol) | vol.97 |
| Number (no) | 532 |
| Page | pp.pp.- |
| #Pages | 5 |
| Date of Issue | |