Presentation 2022-05-27
Study on semiconductor-based nonlinear dynamic node for physical reservoir computing system
Seiya Kasai, Shunsuke Saito,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Reservoir computing is a framework of the neural network-based machine learning for time series. In this system, design and control of the network is not necessary, furthermore, the variation of the elements in the network can enhance the computing performance, which allows to use various materials and devices for physical implementation. On the other hand, semiconductor electronics has been seldomly considered for the reservoir computing, although it is the most feasible way not only for practical use but also for ideal physical platform. In this report, for the implementation of the physical reservoir computing system, we investigate the network node using semiconductor tunnel diodes that has spontaneous oscillation dynamics and their connection to modulate the dynamics.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Reservoir Computing / Semiconductor / Nonlinear / Tunnel Diode / Oscillator Node
Paper # ED2022-12,CPM2022-6,SDM2022-19
Date of Issue 2022-05-20 (ED, CPM, SDM)

Conference Information
Committee SDM / ED / CPM
Conference Date 2022/5/27(1days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English) Functional Device Materials, Fabrication, Characterization, and Related Technology
Chair Hiroshige Hirano(TowerPartners Semiconductor) / Hiroki Fujishiro(Tokyo Univ. of Science) / Yuichi Nakamura(Toyohashi Univ. of Tech.)
Vice Chair Shunichiro Ohmi(Tokyo Inst. of Tech.) / Seiya Sakai(Hokkaido Univ.) / Hideki Nakazawa(Hirosaki Univ.)
Secretary Shunichiro Ohmi(AIST) / Seiya Sakai(Nihon Univ.) / Hideki Nakazawa(Fjitsu Lab.)
Assistant Taiji Noda(Panasonic) / Tomoyuki Suwa(Tohoku Univ.) / Ryota Isonoi(SCIOCS) / Yoshitugu Yamamoto(Mitsubishi Electric) / Yasuo Kimura(Tokyo Univ. of Tech.) / Fumihiko Hirose(Yamagata Univ.) / Noriko Bamba(Shinshu Univ.)

Paper Information
Registration To Technical Committee on Silicon Device and Materials / Technical Committee on Electron Devices / Technical Committee on Component Parts and Materials
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Study on semiconductor-based nonlinear dynamic node for physical reservoir computing system
Sub Title (in English)
Keyword(1) Reservoir Computing
Keyword(2) Semiconductor
Keyword(3) Nonlinear
Keyword(4) Tunnel Diode
Keyword(5) Oscillator Node
1st Author's Name Seiya Kasai
1st Author's Affiliation Hokkaido University(Hokkaido Univ.)
2nd Author's Name Shunsuke Saito
2nd Author's Affiliation Hokkaido University(Hokkaido Univ.)
Date 2022-05-27
Paper # ED2022-12,CPM2022-6,SDM2022-19
Volume (vol) vol.122
Number (no) ED-52,CPM-53,SDM-54
Page pp.pp.21-24(ED), pp.21-24(CPM), pp.21-24(SDM),
#Pages 4
Date of Issue 2022-05-20 (ED, CPM, SDM)