Presentation 2012/1/12
A study of feature space for curve fitting
Jun FUJIKI, Shotaro AKAHO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In order to fit a curve to points in a plane, we can apply the hypersurface fitting method to the points mapped to a high dimensional feature space. This is equivalent to the kernel minor component analysis, which solves the same eigenvalue problem as kernel principal component analysis and extracts the eigen vector corresponding to the least eigenvalue. However, the solution tends to be unstable as the dimensionality of the feature space increases, because the representer theorem does not hold for kernel minor component analysis. Another approach is to obtain a curve based on the principal components derived from kernel principal component analysis. However, since the mapped image of the whole input space forms a low dimensional manifold, the input point corresponding to the principal component does not always exist and it is necessary to approximate it by preimage whose property has not been enough understood. This report examines and discusses such behavior.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) feature space / curve fitting / kernel PCA / Hilbert space
Paper # MVE2011-64,PRMU2011-155
Date of Issue

Conference Information
Committee MVE
Conference Date 2012/1/12(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Media Experience and Virtual Environment (MVE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A study of feature space for curve fitting
Sub Title (in English)
Keyword(1) feature space
Keyword(2) curve fitting
Keyword(3) kernel PCA
Keyword(4) Hilbert space
1st Author's Name Jun FUJIKI
1st Author's Affiliation The National Institute of Advanced Industrial Science and Technology (AIST)()
2nd Author's Name Shotaro AKAHO
2nd Author's Affiliation The National Institute of Advanced Industrial Science and Technology (AIST)
Date 2012/1/12
Paper # MVE2011-64,PRMU2011-155
Volume (vol) vol.111
Number (no) 380
Page pp.pp.-
#Pages 6
Date of Issue