| Presentation | 1996/2/3 Performance of fluctuation-driven learning for feedforward neural networks Yuuki KAKUTA, Mitsuaki MITANI, Takahumi OOHORI, Kazuhisa WATANABE, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We propose fluctuation-driven learning method for feedforward neural networks. As each neuron within the network has a random-fluctuation source, the state of each neuron and the error of network outputs fluctuate. This method can make the networks learn by simple averaging processes whether the neuron model is continuous or discrete. Simulations of several tasks demonstrate the validity of the proposed learning method. For example, simulations show that the networks including discrete neurons easily realize the approximation of discontinuous function and the vector quantization. Furthermor, yhey clarify that the networks can learn by pure chaos fluctuations, too. This method can be considered familiar to parallel distributed processing because the modification of every connection can be excuted simply, individually and simultaneously. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | fluctuation-driven learning / feedforward neural network / theory / simulation / multi-layered neural network |
| Paper # | NC95-106 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 1996/2/3(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) | Performance of fluctuation-driven learning for feedforward neural networks |
| Sub Title (in English) | |
| Keyword(1) | fluctuation-driven learning |
| Keyword(2) | feedforward neural network |
| Keyword(3) | theory |
| Keyword(4) | simulation |
| Keyword(5) | multi-layered neural network |
| 1st Author's Name | Yuuki KAKUTA |
| 1st Author's Affiliation | Department of Electrical Engineering, Hokkaido Institute of Technology() |
| 2nd Author's Name | Mitsuaki MITANI |
| 2nd Author's Affiliation | Department of Electrical Engineering, Hokkaido Institute of Technology |
| 3rd Author's Name | Takahumi OOHORI |
| 3rd Author's Affiliation | Department of Electrical Engineering, Hokkaido Institute of Technology |
| 4th Author's Name | Kazuhisa WATANABE |
| 4th Author's Affiliation | Department of Electrical Engineering, Hokkaido Institute of Technology |
| Date | 1996/2/3 |
| Paper # | NC95-106 |
| Volume (vol) | vol.95 |
| Number (no) | 506 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |