Presentation | 2020-10-29 Unsupervised learning based on local interactions between reservoir and readout neurons Tstuki Kato, Satoshi Moriya, Hideaki Yamamoto, Masao Sakuraba, Shigeo Sato, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Reservoir computing is suitable for implementations in edge computing devices thanks to its low computational cost and ease of physical implementation. However, widely used learning methods, such as Ridge regression, require global information of reservoir states and readout layer. This makes it difficult to implement reservoir computing systems in edge computing devices based on analog devices and circuits. Spike-timing-dependent plasticity (STDP) is a learning method based solely on local information. In this study, we propose a reservoir computing model which can classify temporal patterns whose timescale is far longer than that of a reservoir. In order to obtain this property, we implement multiple conduction delays to readout synapses. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | STDP / reservoir computing / conduction delay / spiking neural network |
Paper # | NC2020-12 |
Date of Issue | 2020-10-22 (NC) |
Conference Information | |
Committee | MBE / NC / NLP / CAS |
---|---|
Conference Date | 2020/10/29(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | ME,NC,CAS,NLP |
Chair | Takashi Watanabe(Tohoku Univ.) / Kazuyuki Samejima(Tamagawa Univ) / Kiyohisa Natsume(Kyushu Inst. of Tech.) / Yasuhiro Takashima(Univ. of Kitakyushu) |
Vice Chair | Ryuhei Okuno(Setsunan Univ.) / Rieko Osu(Waseda Univ.) / Takuji Kosaka(Chukyo Univ.) / Hiroki Sato(Sony LSI Design) |
Secretary | Ryuhei Okuno(Akita-noken) / Rieko Osu(NTT) / Takuji Kosaka(ATR) / Hiroki Sato(Kyushu Inst. of Tech.) |
Assistant | Akihiro Karashima(Tohoku Inst. of Tech.) / Jun Akazawa(Meiji Univ. of Integrative Medicine) / Ken Takiyama(TUAT) / Nobuhiko Wagatsuma(Toho Univ.) / Toshikaza Samura(Yamaguchi Univ.) / Hideyuki Kato(Oita Univ.) / Motoi Yamaguchi(TECHNOPRO) / Yohei Nakamura(Hitachi) |
Paper Information | |
Registration To | Technical Committee on ME and Bio Cybernetics / Technical Committee on Neurocomputing / Technical Committee on Nonlinear Problems / Technical Committee on Circuits and Systems |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Unsupervised learning based on local interactions between reservoir and readout neurons |
Sub Title (in English) | |
Keyword(1) | STDP |
Keyword(2) | reservoir computing |
Keyword(3) | conduction delay |
Keyword(4) | spiking neural network |
1st Author's Name | Tstuki Kato |
1st Author's Affiliation | Tohoku University(Tohoku Univ.) |
2nd Author's Name | Satoshi Moriya |
2nd Author's Affiliation | Tohoku University(Tohoku Univ.) |
3rd Author's Name | Hideaki Yamamoto |
3rd Author's Affiliation | Tohoku University(Tohoku Univ.) |
4th Author's Name | Masao Sakuraba |
4th Author's Affiliation | Tohoku University(Tohoku Univ.) |
5th Author's Name | Shigeo Sato |
5th Author's Affiliation | Tohoku University(Tohoku Univ.) |
Date | 2020-10-29 |
Paper # | NC2020-12 |
Volume (vol) | vol.120 |
Number (no) | NC-216 |
Page | pp.pp.21-23(NC), |
#Pages | 3 |
Date of Issue | 2020-10-22 (NC) |