| 講演名 | 2020-03-06 Modular Reservoir Network for Pattern Recognition 戴 一凡(東北大), ?庭 政夫(東北大), 佐藤 茂雄(東北大), |
|---|---|
| PDFダウンロードページ |  PDFダウンロードページへ |
| 抄録(和) | |
| 抄録(英) | This work is based on liquid state machine (LSM) [1], which is a reservoir network [2] that yields deep relationship with neurophysiology and has possessed great success in time series processing. However, compared with popular feedforward artificial neural networks, the simulation cost for LSM involving spiking neuron is high, which limits the size and functionality of LSM. For better application of LSM, the purpose of this work is to 1) scale up the reservoir efficiently with low computational cost and 2) boost up the functionality by changing topology and synapses of the network. Firstly, we introduce the modular structure in which the synapses of neurons in different modules form directed-acyclic-graph while the neurons in the same modules are connected recurrently. Under this topology, a divide-and-conquer based algorithm could be applied to reduce the computational complexity dramatically. Secondly, we integrate the study on visual cortex [3], in which a Hough transform based convincible model of motion detection is invented, to improve the performance of reservoir network in pattern recognition. We use specifically designed input synapses with which the response of post neurons could perform the Hough transform with no extra costs. Experimentally, we proved that such assignment could improve the performance a lot compared with the case of random connected input synapses. Numerical experiment is performed with MNIST dataset in which the images are converted into spiking sequences by Poisson encoding. The proposed structure could achieve highest precision against previously reported networks of similar size with around 1200 neurons (small-world, random and network proposed in [4] with probability of connections decayed by distance). As for the readout function, both SVM and linear map are implemented, in which the SVM readout performs better than linear map most of the time because of the overfitting in linear map. Moreover, the modular based proposed method shows great system robustness. As indicated in [5], even though the reservoir network could resist the input noise, slight system noise (randomly disable some synapses) would deteriorate the performance a lot. We tested the proposed network under system noise from 0.1% to 20%, in all level of noise, the proposed structure shows significant improvement from previously reported network. Besides, we also find that the proposed reservoir network suffers less performance loss compared with CNN when the labeled training data is not enough, which could have many potential applications. |
| キーワード(和) | |
| キーワード(英) | LSMReservoir computingModularPattern recognitionPattern recognitionHough transformVisual cortex |
| 資料番号 | NC2019-110 |
| 発行日 | 2020-02-26 (NC) |
| 研究会情報 | |
| 研究会 | NC / MBE |
|---|---|
| 開催期間 | 2020/3/4(から3日開催) |
| 開催地（和） | 電気通信大学 |
| 開催地（英） | University of Electro Communications |
| テーマ（和） | NC, ME, 一般 |
| テーマ（英） | Neuro Computing, Medical Engineering, etc. |
| 委員長氏名（和） | 庄野 逸(電通大) / 野村 泰伸(阪大) |
| 委員長氏名（英） | Hayaru Shouno(UEC) / Taishin Nomura(Osaka Univ.) |
| 副委員長氏名（和） | 鮫島 和行(玉川大) / 渡邊 高志(東北大) |
| 副委員長氏名（英） | Kazuyuki Samejima(Tamagawa Univ) / Takashi Watanabe(Tohoku Univ.) |
| 幹事氏名（和） | 吉本 潤一郎(奈良先端大) / 安部川 直稔(NTT) / 伊良皆 啓治(九大) |
| 幹事氏名（英） | Junichiro Yoshimoto(NAIST) / Naotoshi Abekawa(NTT) / Keiji Iramina(Kyushu Univ.) |
| 幹事補佐氏名（和） | 篠崎 隆志(NICT) / 瀧山 健(東京農工大) / 鈴木 康之(阪大) / 辛島 彰洋(東北工大) |
| 幹事補佐氏名（英） | Takashi Shinozaki(NICT) / Ken Takiyama(TUAT) / Yasuyuki Suzuki(Osaka Univ.) / Akihiro Karashima(Tohoku Inst. of Tech.) |
| 講演論文情報詳細 | |
| 申込み研究会 | Technical Committee on Neurocomputing / Technical Committee on ME and Bio Cybernetics |
|---|---|
| 本文の言語 | ENG |
| タイトル（和） | |
| サブタイトル（和） | |
| タイトル（英） | Modular Reservoir Network for Pattern Recognition |
| サブタイトル（和） | |
| キーワード(1)（和/英） | / LSMReservoir computingModularPattern recognitionPattern recognitionHough transformVisual cortex |
| 第 1 著者 氏名（和/英） | 戴 一凡 / Yifan Dai |
| 第 1 著者 所属（和/英） | 東北大学(略称：東北大) Tohoku University(略称：Tohoku Univ.) |
| 第 2 著者 氏名（和/英） | ?庭 政夫 / Masao Sakuraba |
| 第 2 著者 所属（和/英） | 東北大学(略称：東北大) Tohoku University(略称：Tohoku Univ.) |
| 第 3 著者 氏名（和/英） | 佐藤 茂雄 / Shigeo Sato |
| 第 3 著者 所属（和/英） | 東北大学(略称：東北大) Tohoku University(略称：Tohoku Univ.) |
| 発表年月日 | 2020-03-06 |
| 資料番号 | NC2019-110 |
| 巻番号（vol） | vol.119 |
| 号番号（no） | NC-453 |
| ページ範囲 | pp.199-199(NC), |
| ページ数 | 1 |
| 発行日 | 2020-02-26 (NC) |