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

Information Enhancement for Interpreting Internal Representation in Neural Networks

Research Project

Project/Area Number 21500221
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeSingle-year Grants
Section一般
Research Field Sensitivity informatics/Soft computing
Research InstitutionTokai University

Principal Investigator

KAMIMRUA Ryotaro  東海大学, 情報教育センター, 教授 (80176643)

Co-Investigator(Kenkyū-buntansha) TAKEUCHI Haruhiko  独立行政法人産業技術総合研究所, ヒューマンライフテクノロジー研究部門, 主任研究員 (00357401)
Project Period (FY) 2009 – 2011
Project Status Completed (Fiscal Year 2011)
Budget Amount *help
¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2011: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2010: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2009: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Keywordsニューラルネットワーク / 自己組織化マップ / 特徴抽出 / 情報理論 / エントロピー / 相互情報量 / エンハンスメント / 内部表現
Research Abstract

In this study, we proposed a new method called "information enhancement" for interpreting the mechanism of neural networks. We tried to interpret all the components and all the possible combinations of the components by this information enhancement. Applied to the self-organizing maps, we found that clear class structure could be produced for interpretation. In addition, we found that the method was related to a new model for neural networks taking into account the social activities of neurons.

Report

(4 results)
  • 2011 Annual Research Report   Final Research Report ( PDF )
  • 2010 Annual Research Report
  • 2009 Annual Research Report
  • Research Products

    (55 results)

All 2012 2011 2010 2009

All Journal Article (55 results) (of which Peer Reviewed: 55 results)

  • [Journal Article] Comprehensibility maximization and humanly comprehensible representations2012

    • Author(s)
      R. Kamimura
    • Journal Title

      International Journal of General Systems

      Volume: Vol.41, No.3 Pages: 265-287

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Self-enhancement learning : target-creating learning and its application to self-organizing maps2011

    • Author(s)
      R. Kamimura
    • Journal Title

      Biological Cybernetics

      Volume: Vol.104 Pages: 305-338

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Constrained Information maximization by free energy minimization2011

    • Author(s)
      R. Kamimura
    • Journal Title

      International Journal of General Systems

      Volume: Vol.40.No.7 Pages: 701-725

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Selective information enhancement learning for creating interpretable representations in competitive learning2011

    • Author(s)
      R. Kamimura
    • Journal Title

      Neural Networks

      Volume: Vol.24, No.4 Pages: 387-405

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Explicit class structure by weighted cooperative learning2011

    • Author(s)
      R. Kamimura
    • Journal Title

      Lecture Notes in Computer Science

      Volume: Vol.6791 Pages: 109-116

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Cooperation control and enhanced class structure in self-organizing maps2011

    • Author(s)
      R. Kamimura
    • Journal Title

      Proceedings of the International Joint Conference on Neural Networks

      Pages: 689-695

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Explicit class structure with closeness and similarity between neurons2011

    • Author(s)
      R. Kamimura
    • Journal Title

      Proceedings of World Congress on Nature and Biologically Inspired Computing(NaBIC)

      Pages: 92-98

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Supposed maximum information for comprehensible representations in SOM2011

    • Author(s)
      R. Kamimura
    • Journal Title

      Neurocomputing

      Volume: Vol.74, No.7 Pages: 1116-1134

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Structural enhanced information and its application to improved visualization of self-organizing maps2011

    • Author(s)
      R. Kamimura
    • Journal Title

      Applied Intelligence

      Volume: Vol.34, No.1 Pages: 102-115

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Self-enhancement learning : target-creating learning and its application to self-organizmg maps2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Biological Cybernetics

      Volume: 104 Issue: 4-5 Pages: 305-338

    • DOI

      10.1007/s00422-011-0434-x

    • Related Report
      2011 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Constrained Information maximization by free energy minimization2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      International Journal of General Systems

      Volume: 40 Issue: 7 Pages: 701-725

    • DOI

      10.1080/03081079.2010.549486

    • Related Report
      2011 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Selective information enhancement learning for creating interpretable representations in competitive learning2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Neural Networks

      Volume: 24 Pages: 387-405

    • Related Report
      2011 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Comprehensibility maximization and humanly comprehensible representations2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      International Journal of General Systems

      Volume: 41 Issue: 3 Pages: 265-287

    • DOI

      10.1080/03081079.2011.643471

    • Related Report
      2011 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Explicit Class Structure by Weighted Cooperative Learning2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      LNCS

      Volume: 6791 Pages: 109-116

    • DOI

      10.1007/978-3-642-21735-7_14

    • ISBN
      9783642217340, 9783642217357
    • Related Report
      2011 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Cooperation control and enhanced class structure in self-organizing maps2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Proceedings of International Joint Conference on Neural Networks

      Pages: 689-695

    • DOI

      10.1109/ijcnn.2011.6033288

    • Related Report
      2011 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Explicit class structure with closeness and similarity between neurons2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Proceedings of NaBIC

      Pages: 92-98

    • DOI

      10.1109/nabic.2011.6089423

    • Related Report
      2011 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Supposed maximum information for comprehensible representations in SOM2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Neurocomputing

      Volume: 74 Pages: 1116-1134

    • Related Report
      2010 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Structural enhanced information and its application to improved visualization of self-organizing maps2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Applied Intelligence

      Volume: 34 Pages: 102-115

    • Related Report
      2010 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Information-theoretic approach to interpret internal representations of self-Organizing maps2011

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Self Organizing Maps-Applications and Novel Algorithm Design

      Pages: 1-30

    • Related Report
      2010 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Information-theoretic enhancement learning and its application to visualization of self-organizing maps2010

    • Author(s)
      R. Kamimura
    • Journal Title

      Neurocomputing

      Volume: Vol.73, No.13-15 Pages: 2642-2664

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Information enhancement for interpreting competitive learning2010

    • Author(s)
      R. Kamimura
    • Journal Title

      International Journal of General Systems

      Volume: Vol.39, No.7 Pages: 705-728

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Pseudo-network growing for gradual interpretation of input patterns2010

    • Author(s)
      R. Kamimura
    • Journal Title

      LNCS(Lecture Notes in Computer Science, Neural Information Processing, Models and Applications)

      Volume: Vol.6444 Pages: 375-382

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Generation of comprehensible representations by supposed maximum information2010

    • Author(s)
      R. Kamimura
    • Journal Title

      LNCS(Lecture Notes in Computer Science, Artificial Neural Networks)

      Volume: Vol.6353 Pages: 333-342

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Information-theoretic competitive and cooperative learning for self-organizing Maps2010

    • Author(s)
      R. Kamimura
    • Journal Title

      LNCS(Lecture Notes in Computer Science, Neural Information Processing, Models and Applications)

      Volume: Vol.6444 Pages: 423-430

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Explicit class structure produced by information-theoretic competitive and cooperative learning2010

    • Author(s)
      R. Kamimura
    • Journal Title

      Proceedings of World Congress on Nature and Biologically Inspired Computing

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Contradiction resolution and its application to self-organizing maps2010

    • Author(s)
      R. Kamimura
    • Journal Title

      Proceedings of IASTED artificial intelligence and applications

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Information maximization for variable selection in competitive learning2010

    • Author(s)
      R. Kamimura
    • Journal Title

      Proceedings of IASTED artificial intelligence and applications

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Information-theoretic enhancement learning and its apphcation to visualization of self-organizing maps2010

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Neurocomputing

      Volume: 73 Pages: 2642-2664

    • Related Report
      2010 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Pseudo-network growing for gradual interpretation of input patterns2010

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      LNCS

      Volume: 6444 Pages: 375-382

    • Related Report
      2010 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Generation of comprehensible representations by supposed maximum information2010

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      LNCS

      Volume: 6353 Pages: 333-342

    • Related Report
      2010 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Information-theoretic competitive and cooperative learning for self-organizing Maps2010

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      LNCS

      Volume: 6444 Pages: 423-430

    • Related Report
      2010 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Explicit class structure produced by information-theoretic competitive and cooperative learning2010

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Proceedings of World Congress on Nature and Biologically Inspired Computing

      Pages: 628-633

    • Related Report
      2010 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Relative relaxation and weighted information loss to simplify and stabilize feature detection2009

    • Author(s)
      R. Kamimura
    • Journal Title

      Journal of Advanced Computational Intelligence and Intelligent Informatics

      Volume: Vol.13, No.4 Pages: 489-498

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Enhancing and relaxing competitive units for feature discovery2009

    • Author(s)
      R. Kamimura
    • Journal Title

      Neural Processing Letters

      Volume: Vol.30, No.1 Pages: 37-57

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] An information-theoretic approach to feature extraction in competitive learning2009

    • Author(s)
      R. Kamimura
    • Journal Title

      Neurocomputing

      Volume: Vol.72, No.10-12 Pages: 2693-2704

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Feature discovery by information loss2009

    • Author(s)
      R. Kamimura
    • Journal Title

      Journal of Computers

      Volume: Vol.4, No.10 Pages: 943-953

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Feature detection by structural enhanced information2009

    • Author(s)
      R. Kamimura
    • Journal Title

      LNCS(Lecture Notes in Computer Science, Advances in Neuro-Information Processing)

      Volume: Vol.5507 Pages: 171-178

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Partially enhanced competitive learning2009

    • Author(s)
      R. Kamimura
    • Journal Title

      LNCS(Lecture Notes in Computer Science, Advances in Neuro-Information Processing)

      Volume: Vol.5507 Pages: 163-170

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Collective activations to generate self-organizing maps2009

    • Author(s)
      R. Kamimura
    • Journal Title

      LNCS(Lecture Notes in Computer Science, Advances in Neuro-Information Processing)

      Volume: Vol.5507 Pages: 943-950

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Enhanced visualization by combing SOM and mixture models2009

    • Author(s)
      R. Kamimura
    • Journal Title

      LNCS(Lecture Notes in Computer Science, Advances in Neuro-Information Processing)

      Volume: Vol.5507 Pages: 268-275

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Self-enhancement learning : self-supervised and target-creating learning2009

    • Author(s)
      R. Kamimura
    • Journal Title

      Proceedings of international joint conference of neural networks

      Pages: 1503-1509

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Self-supervised learning by information enhancement : target-generating and spontaneous learning for competitive learning2009

    • Author(s)
      R. Kamimura
    • Journal Title

      Proceedings of IEEE international conference on systems, man and cybernetics

      Pages: 113-119

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] Structural enhanced information to detect features in competitive learning2009

    • Author(s)
      R. Kamimura
    • Journal Title

      Proceedings of IEEE international conference on systems, man and cybernetics

      Pages: 3395-3401

    • Related Report
      2011 Final Research Report
    • Peer Reviewed
  • [Journal Article] An information-theoretic approach to feature extraction in competitive learning2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Neurocomputing 72(10-12)

      Pages: 2693-2704

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Enhancing and Relaxing Competitive Units for Feature Discovery2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Neural processing letters 30(1)

      Pages: 37-57

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Feature Detection by Structural Enhanced Information2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Lecture Notes in Computer Science 5507

      Pages: 171-178

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Partially Enhanced Competitive Learning2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Lecture Notes in Computer Science 5507

      Pages: 163-170

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Feature Discovery by Information Loss2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Journal of Computers 4(10)

      Pages: 943-953

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Collective Activations to Generate Self-Organizing Maps2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Lecture Notes in Computer Science 5507

      Pages: 943-950

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Enhanced Visualization by Combing SOM and Mixture Models2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Lecture Notes in Computer Science 5507

      Pages: 268-275

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Relative Relaxation and Weighted Information Loss to Simplify and Stabilize Feature Detection2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Journal of Advanced Computational Intelligence and Intelligent Informatics 13(4)

      Pages: 489-498

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Self-enhancement learning : Self-supervised and target-creating learning2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Proceedings of the international joint conference on neural networks

      Pages: 1503-1509

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Selective enhancement learning in competitive learning2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Proceedings of the international joint conference on neural networks

      Pages: 1497-1502

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Self-supervised learning by information enhancement : Target-generating and spontaneous learning for competitive learning2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Proceedings of IEEE international conference on systems, man and cybernetics

      Pages: 113-119

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Structural enhanced information to detect features in competitive learning2009

    • Author(s)
      Ryotaro Kamimura
    • Journal Title

      Proceedings of IEEE international conference on systems, man and cybernetics

      Pages: 3395-3401

    • Related Report
      2009 Annual Research Report
    • Peer Reviewed

URL: 

Published: 2009-04-01   Modified: 2016-04-21  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi