Misc - TANAKA Gouhei
-
Reservoir computing devices that realize pattern information representation and processing directly in physics: Its advantages
廣瀬明, 田中剛平, 武田征士, 山根敏志, 沼田秀俊, 金澤直輝, HEROUX J. B, 中野大樹, 中根了昌
電子情報通信学会技術研究報告 118 ( 470(NC2018 44-88)(Web) ) 117‐120 (WEB ONLY) 2019.02
-
スピン波リザバーコンピューティングチップデバイス
中根了昌, 田中剛平, 廣瀬明
応用物理学会春季学術講演会講演予稿集(CD-ROM) 66th ROMBUNNO.12p‐W933‐10 2019.02
-
「ニューロ的」ハードウエアと物理リザバーコンピューティング
廣瀬明, 武田征士, ヘロー ジャンベノ, 金澤直輝, 沼田英俊, 山根敏志, 中野大樹, 中根了昌, 田中剛平
29th 2019
-
Physical rsssrvoir computing using lightwavss
廣瀬明, 武田征士, HEROUX Jean Benoit, 沼田英俊, 金澤直輝, 山根敏行, 中野大樹, 中根了昌, 田中剛平
Optics & Photonics Japan講演予稿集(CD-ROM) 2019 2019
-
スピン波を用いた機械学習チップデバイス
ナカネ リョウショウ, タナカ ゴウヘイ, ヒロセ アキラ
14 ( 6 ) 329 - 334 2019
-
スピン波リザバーコンピューティングチップにおける実空間情報分布
市村剛大, 中根了昌, 田中剛平, 廣瀬明, 中根了昌, 田中剛平, 廣瀬明
29th 2019
-
招待講演 複雑ネットワークの動的頑強性
タナカ ゴウヘイ
電子情報通信学会技術研究報告 = IEICE technical report : 信学技報 118 ( 316 ) 33 - 38 2018.11
-
Dynamical robustness of complex networks
田中剛平
電子情報通信学会技術研究報告 118 ( 316(CCS2018 33-45) ) 33‐38 2018.11
-
Nonlinear Time Series Prediction using Multi‐Step Learning Echo State Networks
秋山貴則, 田中剛平, 田中剛平
電子情報通信学会技術研究報告 118 ( 284(IBISML2018 44-104)(Web) ) 293‐299 (WEB ONLY) 2018.10
-
ハードウェアで実現するリザバーコンピューティング
中野大樹, 沼田英俊, 山根敏志, 武田征士, HEROUX Jean Benoit, 金澤直樹, 田中剛平, 中根了昌, 廣瀬明
電気学会全国大会講演論文集(CD-ROM) 2018 ROMBUNNO.S16‐4 2018.03
-
大規模停電を想定した電力系統モデルにおけるカスケード故障
117 ( 121 ) 63 - 66 2017.07
-
リザーバ・コンピューティングにおける複素ニューラルネットワーク
廣瀬明, 武田征士, 田中剛平, 中根了昌, 山根敏志, 中野大樹, 中川茂
2017 2017
-
Spontaneous Oscillation in coupled system composed of excitable units
Morino Kai, Tanaka Gouhei, Aihara Kazuyuki
Meeting Abstracts of the Physical Society of Japan 72.1 ( 0 ) 3035 - 3035 2017
Language：Japanese Publisher：The Physical Society of Japan
<p>今回の発表では，興奮性を示す位相振動子を結合した系において，結合強度を強化することで起こる自発的振動現象の解析結果について報告する．今回の系では結合強度を上げることで安定平衡点がサドルノード分岐により消失して自発的な振動が生じるが，全ての素子が振動を開始する場合と一部の素子のみが振動を開始する場合がパラメータにより分かれる．この振動開始のメカニズムを分岐解析によって詳細に解析した結果を報告する．</p>
-
外部フィードバック・レーザを用いたリザーバコンピューティング
武田征士, 山根敏志, 中野大樹, 田中剛平, 中根了昌, 廣瀬明, 中川茂
2017 2017
-
ニューロン特性の不均一性を考慮したエコーステートネットワーク
田中剛平, 中根了昌, 山根敏志, 中野大樹, 武田征士, 中川茂, 廣瀬明
2017 2017
-
Dynamics of Cellular Systems and Bifurcation Theory
TANAKA Gouhei
Seibutsu Butsuri 56 ( 6 ) 340 - 344 2016
Language：Japanese Publisher：The Biophysical Society of Japan General Incorporated Association
-
A pruning method based on weight variation information for feedforward neural networks
Zhiqiang Tong, Gouhei Tanaka
IFAC-PapersOnLine 28 ( 18 ) 221 - 226 2015.11
Language：English
Artificial neural networks are powerful tools for many information processing tasks such as pattern recognition, data mining, optimization, and prediction. It is a significant problem to find optimal structures of artificial neural networks for drawing out their high computational performance. Downsizing of network structure is also an issue to be considered for hardware implementation of largescale neural networks. In this study, we propose a pruning method to find a compact structure of feedforward neural networks with high generalization ability in classification problems. Our method evaluates the significance of neuron nodes using the information on weight variation in the training process and prune the insignificant nodes preferentially unless the classification accuracy is degraded. Numerical experiments with several benchmark datasets show that the proposed method is effective compared with other methods.
-
A hybrid model for hepatitis B virus
Ben-Gong Zhang, Gouhei Tanaka, Kazuyuki Aihara, Masao Honda, Shuichi Kaneko, Luonan Chen
IFAC-PapersOnLine 28 ( 18 ) 37 - 40 2015.11
Language：English
The hepatitis B virus (HBV) is a kind of pandemic infectious disease. It is becoming a major health problem in Asian and African countries, however, less so in other areas around the world. There is a vast background for studying HBV. However, some problems still require further study. For example, when the patient is infected by HBV, how to arrange the treatment schedule? Using long term continuous treatment or others? Therefore, we wish to readdress the study of HBV. First, we propose a new hybrid model that contains both continuous and discrete variables. Second, we analyze its dynamical behavior. We hope that this study may provide some new insight for HBV disease and some new guidance for the therapy.
-
Dynamical robustness of complex biological networks
Gouhei Tanaka, Kai Morino, Kazuyuki Aihara
Mathematical Approaches to Biological Systems: Networks, Oscillations, and Collective Motions 29 - 53 2015.01
Language：English Publisher：Springer Japan
Dynamical behavior of biological systems is maintained by interactions between biological units such as neurons, cells, proteins, and molecules. It is a challenging issue to understand robustness of biological interaction networks from a viewpoint of dynamical systems. In this chapter, we introduce the concept of dynamical robustness in complex networks and demonstrate its application to biological networks. First, we introduce the framework for studying the dynamical robustness through analyses of coupled Stuart-Landau oscillators with various types of network structures. Second, based on the framework, we examine the dynamical robustness of neuronal firing activity in networks of synaptically coupled Morris- Lecar neuron models. Our analyses suggest that a consideration of both network structure and dynamics is crucial in elucidating biological robustness.
