| Presentation | 2001/5/18 A Synthesis and Learning Method for Multilayer Neural Networks Embedding Gate Elements and with Multidimensional Inputs Makoto Kourin, Kenji Nakayama, Akihiro Hirano, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this paper, a synthesis and learning method for the neural network with embedded gate units and having multi-dimensional input is proposed. In multi-dimensional input, gate functions are controlled in multi-dimensional space. In this case, the direction of the gate function should be considered. Furthermore, the switching points should be considered on the u_j axis, which is the input potential of the hidden units. The initialization and control methods for gate functions including the direction, the switching point and te inclination, are proposed. The stabilization methods, which were proposed, are modified taking multi-dimensional input into account. The gate functions can be trained together with the connection weights. Discontinuous function approximation is demonstrated to confirm usefulness of the proposed method. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Neural networks / Modular networks / Gates / Sigmoid functions / Learning / Discontinuous functions |
| Paper # | NC2001-6 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 2001/5/18(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 | ENG |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A Synthesis and Learning Method for Multilayer Neural Networks Embedding Gate Elements and with Multidimensional Inputs |
| Sub Title (in English) | |
| Keyword(1) | Neural networks |
| Keyword(2) | Modular networks |
| Keyword(3) | Gates |
| Keyword(4) | Sigmoid functions |
| Keyword(5) | Learning |
| Keyword(6) | Discontinuous functions |
| 1st Author's Name | Makoto Kourin |
| 1st Author's Affiliation | Division of Electronics and Computer Science, Graduate School of Natural Science and Technology Kanazawa University() |
| 2nd Author's Name | Kenji Nakayama |
| 2nd Author's Affiliation | Dept. of Information and Systems Eng. Faculty of Eng. Kanazawa University |
| 3rd Author's Name | Akihiro Hirano |
| 3rd Author's Affiliation | Dept. of Information and Systems Eng. Faculty of Eng. Kanazawa University |
| Date | 2001/5/18 |
| Paper # | NC2001-6 |
| Volume (vol) | vol.101 |
| Number (no) | 94 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |