Learning Curves of Polynomial Kernel Classifiers

Kazushi Ikeda

doi:10.11499/sicep.2004.0.16.1

SICE Annual Conference 2004

DOI https://doi.org/10.11499/sicep.2004.0.16.1

Conference information

Host: The Society of Instrument and Control Engineers

Co-host: The IEEE Industrial Electronics Society, The IEEE Robotics and Automation Society, The IEEE Control Systems Society, The IEEE Systems, Man and Cybernetics Society, The Instrumentation, Systems, and Automation Society

Learning Curves of Polynomial Kernel Classifiers

*Kazushi Ikeda

Author information

Keywords: learning theory, generalization error, kernel method

CONFERENCE PROCEEDINGS FREE ACCESS

Pages 16

Details

Abstract

This paper derives the learning curve of a polynomial kernel classifier in the asymptotic limit which linearly discriminates vectors nonlinearly mapped from input vectors. It is similar to but smaller than that of a linear dichotomy depending on the relationship between the subset in the feature space on which the mapped vectors lie and the true separating hyperplane.

Corresponding author

Register with J-STAGE for free!