Analysis of On-Line Learning when a Moving Teacher Goes around a True Teacher

Access this Article

Search this Article

Author(s)

Abstract

In the framework of on-line learning, a learning machine might move around a teacher due to the differences in structures or output functions between the teacher and the learning machine or due to noises. The generalization performance of a new student supervised by a moving machine has been analyzed. A model composed of a fixed true teacher, a moving teacher and a student that are all linear perceptrons with noises has been treated analytically using statistical mechanics. It has been proven that the generalization errors of a student can be smaller than that of a moving teacher, even if the student only uses examples from the moving teacher.

Journal

  • Journal of the Physical Society of Japan

    Journal of the Physical Society of Japan 75(2), "24003-1"-"24003-6", 2006-02-15

    The Physical Society of Japan (JPS)

References:  12

  • <no title>

    SAAD D. ed.

    On-Line Learning in Neural Networks, 1998

    Cited by (1)

  • <no title>

    MIYOSHI S.

    Phys. Rev. E 71, 036116, 2005

    Cited by (1)

  • <no title>

    FREUND Y.

    J. Jpn. Soc. Artif. Intell. 14, 771, 1999

    Cited by (1)

  • <no title>

    http://www.boosting.org/

    Cited by (1)

  • <no title>

    NISHIMORI H.

    Statistical Physics of Spin Glasses and Information Processing : An Introduction, 2001

    Cited by (1)

  • <no title>

    MIYOSHI S.

    Proc. 7th Workshop on Information-Based Induction Sciences, 2004, 2004

    Cited by (1)

  • <no title>

    MIYOSHI S.

    IEICE Tech. Rep. NC2004-214, 123, 2005

    Cited by (1)

  • <no title>

    INOUE J. I.

    cond-mat/9708096

    Cited by (1)

  • <no title>

    KROGH A.

    Phys. Rev. E 55, 811, 1997

    Cited by (15)

  • <no title>

    URBANCZIK R.

    Phys. Rev. E 62, 1448, 2000

    Cited by (11)

  • <no title>

    INOUE J.

    Phys. Rev. E 55, 4544, 1997

    Cited by (9)

  • Ensemble Learning of Linear Perceptrons : On-Line Learning Theory

    HARA Kazuyuki , OKADA Masato

    Journal of the Physical Society of Japan 74(11), 2966-2972, 2005-11-15

    References (9) Cited by (11)

Cited by:  5

Codes

  • NII Article ID (NAID)
    110004086865
  • NII NACSIS-CAT ID (NCID)
    AA00704814
  • Text Lang
    ENG
  • Article Type
    Journal Article
  • ISSN
    00319015
  • NDL Article ID
    7813093
  • NDL Source Classification
    ZM35(科学技術--物理学)
  • NDL Call No.
    Z53-A404
  • Data Source
    CJP  CJPref  NDL  NII-ELS  J-STAGE 
Page Top