Information-Theoretic Competitive Learning and its Application to Human Information Processing

2008 Fiscal Year Final Research Report

• Project/Area Number: 18500179
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Sensitivity informatics/Soft computing
• Research Institution: Tokai University
• Principal Investigator: KAMIMURA Ryotaro Tokai University, 情報教育センター, 教授 (80176643)
• Co-Investigator(Kenkyū-buntansha): UCHIDA Osamu 東海大学, 情報理工学部, 講師 (50329306) ; TAKEUCHI Haruhiko 産業技術総合研究所, サービス工学研究センター, 主任研究員 (00357401)
• Project Period (FY): 2006 – 2008
• Keywords: 競合学習 / ニューラルネットワーク / 相互情報量

Research Abstract

競合学習を実現する情報量最大化法を提案し、各種の応用を通して、人間・生体系の情報処理過程の多くが情報量最大化のプロセスであるということを実証しようとした。研究では、情報量最大化学習の高速化の研究をおこない大規模問題への応用可能性を示した。さらに、個別特徴の抽出の方法を提案し、入力パターンの詳細な特徴を抽出することに成功した。

Research Products

• [Journal Article] Forced information for information-theoretic competitive learning (2009)
  • Author(s): R. Kamimura
  • Journal Title: in Machine learning IN-TECH Pages: 125-144
  • Peer Reviewed
• [Journal Article] Assimilation of individual activities to collective ones to produce explicit self-organizing maps (2009)
  • Author(s): Kamimura
  • Journal Title: in Proceedings of the IASTED international conference on artificial intelligence and applications (AIA2009) Pages: 158-162
  • Peer Reviewed
• [Journal Article] Supervised enhanced learning to simplify internal representations of multilayered networks (2009)
  • Author(s): R. Kamimura
  • Journal Title: in Proceedings of the IASTED international conference on artificial intelligence and applications (AIA2009) Pages: 169-174
  • Peer Reviewed
• [Journal Article] Conditional information and information loss for flexible feature extraction (2008)
  • Author(s): R. Kamimura
  • Journal Title: in Proceedings of the international joint conference on neural networks(IJCNN2008) Pages: 2047-2083
  • Peer Reviewed
• [Journal Article] Feature discovery by enhancement and relaxation of competitive units (2008)
  • Author(s): R. Kamimura
  • Journal Title: in Intelligent data engineering and automated learning-IDEAL2008(LNCS), Springer vol.LNCS5326 Pages: 148-155
  • Peer Reviewed
• [Journal Article] Mutual information maximization by free energy-based competitive learning for self-organizing maps (2008)
  • Author(s): R. Kamimura
  • Journal Title: in Proceedings of the international conference on systems, man, and cybernetics(SMC2008) Pages: 1819-1825
  • Peer Reviewed
• [Journal Article] Interpreting and improving multi-layered networks by free energy-based competitive learning (2008)
  • Author(s): R. Kamimura
  • Journal Title: in Proceedings of the international conference on systems, man, and cybernetics(SMC2008) Pages: 1812-1818
  • Peer Reviewed
• [Journal Article] Feature detection and information loss in competitive learning (2008)
  • Author(s): R. Kamimura
  • Journal Title: in Proceedings of the international conference on soft computing and intelligent systems and the international symposium on advanced intelligent systems(SCIS and ISIS2008) Pages: 1144-1148
  • Peer Reviewed
• [Journal Article] Free Energy-based Competitive Learning for Self-Organizing Maps (2008)
  • Author(s): R. Kamimura
  • Journal Title: Proceedings of Artificial Intelligence and Applications Pages: 414-419
  • Peer Reviewed
• [Journal Article] An Information-Theoretic Approach to Feature Extraction in Competitive Learning (2008)
  • Author(s): R.Kamimura, T. Taniguchi and R. Kitajima
  • Journal Title: Proceeding of Artificial Intelligence and Applications Pages: 388-394
  • Peer Reviewed
• [Journal Article] Free Energy-based Competitive Learning and Minimum Information Production Learning (2008)
  • Author(s): R. Kamimura
  • Journal Title: Proceeding of Artificial Intelligence and Applications Pages: 395-400
  • Peer Reviewed
• [Journal Article] Free Energy-Based Competitive Learning for Mutual Information Maximization (2008)
  • Author(s): R. Kamimura
  • Journal Title: Proceedings of IEEE SMC International Conference on Distributed Human-Machine Systems Pages: 215-222, 414-419
  • Peer Reviewed
• [Journal Article] Supervised Competitive Learning for Explicit Internal Representations (2008)
  • Author(s): R. Kamimura
  • Journal Title: Proceedings of IEEE SMC International Conference on Distributed Human-Machine Systems Pages: 223-227
  • Peer Reviewed
• [Journal Article] Greedy network-growing algorithm with Minkowski distances (2007)
  • Author(s): R. Kamimura, O. Uchida and S. Hashimoto
  • Journal Title: International Journal of General Sytems vol.36 Pages: 157-177
  • Peer Reviewed
• [Journal Article] Combining Hard and Soft Competition in Information-Theoretic Learning (2007)
  • Author(s): R. Kamimura
  • Journal Title: Proceedings of IEEE symposium on foundations of computational intelligence Pages: 578-582
  • Peer Reviewed
• [Journal Article] nformation-Theoretic Variable Selection in Neural Networks (2007)
  • Author(s): R. Kamimura, F. Yoshida, T. Yamashita and R. Kitajima
  • Journal Title: Proceedings of IEEE symposium on foundations of computational intelligence Pages: 578-582
  • Peer Reviewed
• [Journal Article] Forced Information and Information Loss for a Student Survey Analysis (2007)
  • Author(s): R. Kamimura
  • Journal Title: Foundations of Computational Intelligence, 2007. FOCI 2007. IEEE Symposium on Pages: 630-636
  • Peer Reviewed
• [Journal Article] Controlled Competitive Learning : Extending Competitive Learning to Supervised Learning (2007)
  • Author(s): R. Kamimura
  • Journal Title: IJCNN 2007. International Joint Conference on 12-17 Aug. 2007 Pages: 1767-1773
  • Peer Reviewed
• [Journal Article] Forced Information Maximization to Accelerate Information-Theoretic Competitive Learning (2007)
  • Author(s): R. Kamimura, and R. Kitajima
  • Journal Title: IJCNN 2007. International Joint Conference on 12-17 Aug. 2007 Pages: 1779-1784
  • Peer Reviewed
• [Journal Article] Partially Activated Neural Networks by Controlling Information (2007)
  • Author(s): R. Kamimura
  • Journal Title: Lecture Notes in Computer Science LNCS4668 Pages: 1611-3349
  • Peer Reviewed
• [Journal Article] Information loss to extract distinctive features in competitive learning (2007)
  • Author(s): R. Kamimura
  • Journal Title: Proceedingsof IEEE Conference on Systems, Man, and Cybernetics Pages: 1217-1222
  • Peer Reviewed
• [Journal Article] Forced Information and Information Loss in information-Theoretic Competitive Learning (2007)
  • Author(s): R. Kamimura
  • Journal Title: Proceedings of Conferences on Artificial Intelligence and Applications Pages: 69-74
  • Peer Reviewed
• [Journal Article] Interpreting Cabinet Approval Ratings by Neural Networks (2007)
  • Author(s): R. Kamimura, F. Yoshida, R. Kitajima
  • Journal Title: Proceedings of Conferences on Artificial Intelligenceand Applications Pages: 64-68
  • Peer Reviewed
• [Journal Article] Cooperative information maximization with Gaussian acitivation functions for self-organizing maps (2007)
  • Author(s): R. Kamimura
  • Journal Title: IEEE Transactions on Neural Networks Vol.17, No.4 Pages: 909-918
  • Peer Reviewed
• [Journal Article] Supervised Information Maximization by Weighted Distance (2007)
  • Author(s): R. Kamimura
  • Journal Title: Proceedings of International Conference on Neural Networks Pages: 3456-3462
  • Peer Reviewed
• [Journal Article] Student Survey by Information-Theoretic Competitive Learning (2006)
  • Author(s): R. Kamimura
  • Journal Title: Proceedingsof IEEE Conference on Systems, Man, and Cybernetics Pages: 5135-5140
  • Peer Reviewed
• [Journal Article] Controlling Excessive Information by Surface Information Criterion for Information-Theoretic Self-Organizing Maps (2006)
  • Author(s): R. Kamimura
  • Journal Title: Proceedingsof IEEE Conference on Systems, Man, and Cybernetics Pages: 5130-5134
  • Peer Reviewed
• [Journal Article] Automatic Inference of Cabinet Approval Ratings by Information-Theoretic Competitive Learning (2006)
  • Author(s): R. Kamimura
  • Journal Title: LNCS 4232 Pages: 897-908
  • Peer Reviewed
• [Journal Article] Self-organizing by Information Maximization : Realizing Self-Organizing Maps by Information-Theoretic Competitive Learning (2006)
  • Author(s): R. Kamimura
  • Journal Title: LNCS 4232 Pages: 925-934
  • Peer Reviewed
• [Journal Article] Collective Information-Theoretic Competitive Learning : Emergency of Improved Performance by Collectively Treated Neurons Learning (2006)
  • Author(s): R. Kamimura, F. Yoshida, R. Kitajima
  • Journal Title: LNCS 4232 Pages: 626-633
  • Peer Reviewed
• [Journal Article] Inference of Cabinet Approval Ratings by Neural Networks (2006)
  • Author(s): R. Kamimura, F. Yoshida, R. Kitajima
  • Journal Title: Proceedings of SCIS Pages: 213-217
  • Peer Reviewed
• [Presentation] Partially enhanced competitive learning (2008)
  • Author(s): R. Kamimura
  • Organizer: in Proceedings of the international conference on neural information processing (ICONIP2008)
  • Place of Presentation: Auckland
  • Year and Date: 20081100
• [Presentation] Feature detection by structural enhanced information (2008)
  • Author(s): R. Kamimura
  • Organizer: in Proceedings of the international conference on neural information processing (ICONIP2008)
  • Place of Presentation: Auckland
  • Year and Date: 20081100
• [Presentation] Enhanced visualization by combining som and mixture models (2008)
  • Author(s): R. Kamimura
  • Organizer: in Proceedings of the international conference on neural information processing (ICONIP2008)
  • Place of Presentation: Auckland
  • Year and Date: 20081100
• [Presentation] Collective activations to generate self-organizing maps (2008)
  • Author(s): R. Kamimura
  • Organizer: in Proceedings of the international conference on neural information processing (ICONIP2008)
  • Place of Presentation: Auckland
  • Year and Date: 20081100
• [Book] 社会知能システム入門(上村担当部分、pp.1-5, 53-82, 90-91) (2007)
  • Author(s): 上村龍太郎、吉田文彦、山下俊恵, 他
  • Total Pages: 1-12
  • Publisher: 東海大学出版会
• [Book] ビジュアルデータアナリシス(上村担当部分、pp.83-93) (2007)
  • Author(s): 山下俊恵、上村龍太郎、高橋隆男、橋本政樹, 他
  • Total Pages: 1-102
  • Publisher: 東海大学出版会