• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

recise analysis and evolutionary synthesis of sparse binary neural networks

Research Project

Project/Area Number 18K11480
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 61040:Soft computing-related
Research InstitutionHosei University

Principal Investigator

Saito Toshimichi  法政大学, 理工学部, 教授 (30178496)

Project Period (FY) 2018-04-01 – 2021-03-31
Project Status Completed (Fiscal Year 2020)
Budget Amount *help
¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000)
Fiscal Year 2020: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2019: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2018: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Keywordsニューラルネットワーク / 周期軌道 / 安定性 / ニューラルネット / バイナリーニューラルネット
Outline of Final Research Achievements

We have studied analysis and synthesis of dynamic binary neural networks characterized by ternary connection parameters and the signum activation function. First, in simple sparse networks (consisting of neurons from three inputs to one output), we have analyzed stability of rotation-type periodic orbits. Based on the analysis results, we have constructed a synthesis method that guarantees storage and local stability of the rotation-type periodic orbits. Second, in three-layer networks, we have given a basic theoretical result on storage and stability of desired periodic orbits. Based on the theory, we have constructed a synthesis method that guarantees storage and global stability of any desired periodic orbits. Third, implementing the networks on FPGA board, periodic orbits have been confirmed experimentally. The FPGA based hardware has applied to control of walking patterns in hexapod robots.

Academic Significance and Societal Importance of the Research Achievements

人工ニューラルネットは、科学技術の発展に大きく貢献しているAIの基礎となるシステムである。AIに関する諸分野では、写像型のニューラルネットは盛んに研究されているが、再帰型のニューラルネットワークはそれほど研究されていない。生成する現象の複雑さと、解析の困難さがその理由である。動的バイナリーニューラルネットは、動作解析とハードウエア実装に有利な、簡素な再帰型ニューラルネットワークである。その解析と合成に関する基礎研究の成果は、AI技術のさらなる発展への貢献が期待できる。

Report

(4 results)
  • 2020 Annual Research Report   Final Research Report ( PDF )
  • 2019 Research-status Report
  • 2018 Research-status Report
  • Research Products

    (22 results)

All 2021 2020 2019 2018 Other

All Journal Article (13 results) (of which Peer Reviewed: 12 results,  Open Access: 4 results) Presentation (8 results) (of which Int'l Joint Research: 4 results,  Invited: 1 results) Remarks (1 results)

  • [Journal Article] Analysis of Switched Dynamical Systems in Perspective of Bifurcation and Multiobjective Optimization2021

    • Author(s)
      Ryutaro Fujikawa, Tomoyuki Togawa, Toshimichi Saito
    • Journal Title

      IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

      Volume: E104.A Issue: 2 Pages: 525-531

    • DOI

      10.1587/transfun.2020EAP1033

    • NAID

      130007978893

    • ISSN
      0916-8508, 1745-1337
    • Year and Date
      2021-02-01
    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] A simple digital spiking neural network: Synchronization and spike-train approximation2021

    • Author(s)
      Hiroaki Uchida, Yuya Oishi, Toshimichi Saito
    • Journal Title

      Discrete & Continuous Dynamical Systems - S

      Volume: 14 Issue: 4 Pages: 1479-1494

    • DOI

      10.3934/dcdss.2020374

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Guaranteed storage and stabilization of desired binary periodic orbits in three-layer dynamic binary neural networks2020

    • Author(s)
      Seitaro Koyama, Toshimichi Saito
    • Journal Title

      Neurocomputing

      Volume: 416 Pages: 12-18

    • DOI

      10.1016/j.neucom.2020.01.105

    • Related Report
      2020 Annual Research Report 2019 Research-status Report
    • Peer Reviewed
  • [Journal Article] Multi-Phase Synchronization Phenomena in a Ring-Coupled System of Digital Spiking Neurons2019

    • Author(s)
      Hiroaki Uchida, Toshimichi Saito
    • Journal Title

      IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

      Volume: E102.A Issue: 1 Pages: 235-241

    • DOI

      10.1587/transfun.E102.A.235

    • NAID

      130007541773

    • ISSN
      0916-8508, 1745-1337
    • Year and Date
      2019-01-01
    • Related Report
      2018 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] Theoretical analysis of dynamic binary neural networks with simple sparse connection2019

    • Author(s)
      Aoki Shunsuke、Koyama Seitaro、Saito Toshimichi
    • Journal Title

      Neurocomputing

      Volume: 341 Pages: 149-155

    • DOI

      10.1016/j.neucom.2019.03.015

    • Related Report
      2019 Research-status Report 2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] Analysis of cellular automata governed by simple time-variant rules2019

    • Author(s)
      Ozawa Takahiro、Saito Toshimichi
    • Journal Title

      Nonlinear Theory and Its Applications, IEICE

      Volume: 10 Issue: 4 Pages: 357-365

    • DOI

      10.1587/nolta.10.357

    • NAID

      130007722646

    • ISSN
      2185-4106
    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Journal Article] A simple multiobjective evolutionary algorithm for design of digital maps2019

    • Author(s)
      Arai Hayate、Saito Toshimichi
    • Journal Title

      Nonlinear Theory and Its Applications, IEICE

      Volume: 10 Issue: 3 Pages: 280-288

    • DOI

      10.1587/nolta.10.280

    • NAID

      130007673020

    • ISSN
      2185-4106
    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Journal Article] Sparse Dynamic Binary Neural Networks for Storage and Switching of Binary Periodic Orbits2019

    • Author(s)
      Anzai Shota、Koyama Seitaro、Aoki Shunsuke、Saito Toshimichi
    • Journal Title

      T. Gedeon et al. (Eds.): ICONIP 2019, LNCS 11954

      Volume: 11954 Pages: 536-542

    • DOI

      10.1007/978-3-030-36711-4_45

    • ISBN
      9783030367107, 9783030367114
    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Journal Article] Application of MOEA/D to a Trade-Off Problem Between Maximum Power Point and Stability2019

    • Author(s)
      Togawa Tomoyuki、Kunii Yusuke、Yasukawa Shin、Saito Toshimichi
    • Journal Title

      Proc. IEEE/CEC

      Volume: 1 Pages: 3221-3226

    • DOI

      10.1109/cec.2019.8790043

    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Journal Article] A Multi-Objective Problem in a PSO-based Control System for Maximum Power Point Tracking2019

    • Author(s)
      Yasukawa Shin、Saito Toshimichi
    • Journal Title

      Proc. IEEE/CEC

      Volume: 1 Pages: 2628-2633

    • DOI

      10.1109/cec.2019.8790176

    • Related Report
      2019 Research-status Report
  • [Journal Article] A simple multiobjective evolutionary algorithm for design of digital maps2019

    • Author(s)
      H. Arai and T. Saito
    • Journal Title

      NOLTA, IEICE

      Volume: E10-N

    • NAID

      130007673020

    • Related Report
      2018 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] FPGA based hardware implementation of simple dynamic binary neural networks2018

    • Author(s)
      Shunsuke Aoki, Seitaro Koyama, Toshimichi Saito
    • Journal Title

      L. Cheng et al. (Eds.): ICONIP 2018, LNCS

      Volume: 11307 Pages: 647-655

    • DOI

      10.1007/978-3-030-04239-4_58

    • ISBN
      9783030042387, 9783030042394
    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Journal Article] A ladder-type digital spiking neural network2018

    • Author(s)
      Hiroaki Uchida, Toshimichi Saito
    • Journal Title

      L. Cheng et al. (Eds.): ICONIP 2018, LNCS

      Volume: 11301 Pages: 555-562

    • DOI

      10.1007/978-3-030-04167-0_50

    • ISBN
      9783030041663, 9783030041670
    • Related Report
      2018 Research-status Report
    • Peer Reviewed
  • [Presentation] Connection Sparsification and Orbit Stabilization of Dynamic Binary Neural Networks based on Multiobjective Evolutionary Algorithms2020

    • Author(s)
      Tomoyuki Togawa, Toshimichi Saito
    • Organizer
      IEEE WCCI
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] バイナリーニューラルネットに基づくバイナリーデータの分類2020

    • Author(s)
      阪健人、外川智之、斎藤利通
    • Organizer
      電子情報通信学会技術研究報告、NC2020-8
    • Related Report
      2020 Annual Research Report
  • [Presentation] 時変パラメータを有する動的バイナリーニューラルネット2020

    • Author(s)
      鈴木匠、安西翔太、斎藤利通
    • Organizer
      電子情報通信学会技術研究報告、NC2020-9
    • Related Report
      2020 Annual Research Report
  • [Presentation] 音楽信号生成のための簡素なセルオートマトンの解析2020

    • Author(s)
      宇田川穂高、斎藤利通
    • Organizer
      電子情報通信学会技術研究報告、NLP2020-33
    • Related Report
      2020 Annual Research Report
  • [Presentation] 進化的多目的最適化と非線形回路2019

    • Author(s)
      斉藤 利通
    • Organizer
      電子情報通信学会 非線形問題研究会
    • Related Report
      2019 Research-status Report
    • Invited
  • [Presentation] Evolutionary synthesis of simple digital spike maps2018

    • Author(s)
      Toshimichi Saito
    • Organizer
      The 26th Nonlinear Dynamics of Electronic Systems conference
    • Related Report
      2018 Research-status Report
    • Int'l Joint Research
  • [Presentation] Stability analysis of sparse binary neural networks2018

    • Author(s)
      Seitaro Koyama
    • Organizer
      The 26th Nonlinear Dynamics of Electronic Systems conference
    • Related Report
      2018 Research-status Report
    • Int'l Joint Research
  • [Presentation] Synchronization of spike-trains in a coupled system of digital spiking neurons2018

    • Author(s)
      Horoaki Uchida
    • Organizer
      The 26th Nonlinear Dynamics of Electronic Systems conference
    • Related Report
      2018 Research-status Report
    • Int'l Joint Research
  • [Remarks] 法政大学学術研究データベース

    • URL

      http://kenkyu-web.i.hosei.ac.jp/Profiles/14/0001303/profile.html

    • Related Report
      2018 Research-status Report

URL: 

Published: 2018-04-23   Modified: 2022-01-27  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi