Generalization Ability of Dynamic Systems by Using Second Order Derivatives of Universal Learning Network

• Han Min

  Kyushu University

• Hirasawa Kotaro

  Kyushu University

• Hu Jinglu

  Kyushu University

• Murata Junichi

  Kyushu University

Bibliographic Information

Other Title

• Generalization Ability of Dynamic System by Using Second Order Derivatives of Universal Learning Network

Abstract

This paper studies how the generalization ability of modeling of the dynamic systems can be improved by taking advantages of second order derivatives of the criterion function with respect to the external inputs that can be considered as a direct implementation of the well-known regularization technique. Computation of second order derivatives of Universal Learning Network for a dynamic network are discussed. Simulation studies of a nonlinear dynamic system and a real robot system are carried out to demonstrate the effectiveness of the proposed method. Simulation results show that the proposed method can improve the generalization ability of neural networks sufficiently by selecting an appropriate regularization parameter.

Journal

• IEEJ Transactions on Electronics, Information and Systems

  IEEJ Transactions on Electronics, Information and Systems 119 (5), 567-574, 1999

  The Institute of Electrical Engineers of Japan

Keywords

• Generalization Ability
• Universal Learning Network
• Regularization Technique
• Second Order Derivatives

Details

• CRID: 1390282679588391168
• NII Article ID: 130006846008
• NII Article ID: 10002816632
• NII Book ID: AN10065950
• DOI: 10.1541/ieejeiss1987.119.5_567
• ISSN: 13488155
• ISSN: 03854221
• NDL BIB ID: 4716443
• Web Site: https://ndlsearch.ndl.go.jp/books/R000000004-I4716443
• Web Site: https://www.jstage.jst.go.jp/article/ieejeiss1987/119/5/119_5_567/_pdf
• Text Lang: en
• Data Source:
  • JaLC
  • NDL
  • Crossref
  • CiNii Articles
• Abstract License Flag: Disallowed