| Presentation | 2020-03-05 An extension of the H_infinity learning to deep neural networks Yasuhiro Sugawara, Kiyoshi Nishiyama, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In recent years, deep neural networks have achieved remarkable research results. In this study, we propose a method to extend the H∞-learning proposed by one of the authors to deep neural networks. The H∞-learning is a learning method that addresses the difficulties of learning in neural networks. The H∞-learning is derived from applying the extended H∞ filter to a state space model of neural network including an observation matrix. This study extends the H∞-learning to deep neural networks by only changing the calculation of this observation matrix. And we also derive a method to recursively calculate each element of the matrix. In addition, the learning performance is evaluated by simulation in comparison with the conventional backpropagation method. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | neural network / learning algorithm / H∞-learning / backpropagation |
| Paper # | NC2019-92 |
| Date of Issue | 2020-02-26 (NC) |
| Conference Information | |
| Committee | NC / MBE |
|---|---|
| Conference Date | 2020/3/4(3days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | University of Electro Communications |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Neuro Computing, Medical Engineering, etc. |
| Chair | Hayaru Shouno(UEC) / Taishin Nomura(Osaka Univ.) |
| Vice Chair | Kazuyuki Samejima(Tamagawa Univ) / Takashi Watanabe(Tohoku Univ.) |
| Secretary | Kazuyuki Samejima(NAIST) / Takashi Watanabe(NTT) |
| Assistant | Takashi Shinozaki(NICT) / Ken Takiyama(TUAT) / Yasuyuki Suzuki(Osaka Univ.) / Akihiro Karashima(Tohoku Inst. of Tech.) |
| Paper Information | |
| Registration To | Technical Committee on Neurocomputing / Technical Committee on ME and Bio Cybernetics |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | An extension of the H_infinity learning to deep neural networks |
| Sub Title (in English) | |
| Keyword(1) | neural network |
| Keyword(2) | learning algorithm |
| Keyword(3) | H∞-learning |
| Keyword(4) | backpropagation |
| 1st Author's Name | Yasuhiro Sugawara |
| 1st Author's Affiliation | Iwate University(Iwate University) |
| 2nd Author's Name | Kiyoshi Nishiyama |
| 2nd Author's Affiliation | Iwate University(Iwate University) |
| Date | 2020-03-05 |
| Paper # | NC2019-92 |
| Volume (vol) | vol.119 |
| Number (no) | NC-453 |
| Page | pp.pp.95-100(NC), |
| #Pages | 6 |
| Date of Issue | 2020-02-26 (NC) |