A new method for performing a nonlinear form of Principal Component Analysis is proposed. By the use of integral operator kernel functions, one can efficiently compute principal components in highdimensional feature spaces, related to input space by some nonlinear map; for instance the space of all possible d-pixel products in images. We give the derivation of the method and present experimental results on polynomial feature extraction for pattern recognition.
%0 Book Section
%1 scholkopf1997kernel
%A Schölkopf, Bernhard
%A Smola, Alexander
%A Müller, Klaus-Robert
%B Artificial Neural Networks-ICANN'97
%D 1997
%I Springer
%K PCA kernel_PCA linear_algebra methods support_vector_machines
%P 583--588
%T Kernel principal component analysis
%U http://link.springer.com/chapter/10.1007/BFb0020217
%X A new method for performing a nonlinear form of Principal Component Analysis is proposed. By the use of integral operator kernel functions, one can efficiently compute principal components in highdimensional feature spaces, related to input space by some nonlinear map; for instance the space of all possible d-pixel products in images. We give the derivation of the method and present experimental results on polynomial feature extraction for pattern recognition.
@incollection{scholkopf1997kernel,
abstract = {A new method for performing a nonlinear form of Principal Component Analysis is proposed. By the use of integral operator kernel functions, one can efficiently compute principal components in highdimensional feature spaces, related to input space by some nonlinear map; for instance the space of all possible d-pixel products in images. We give the derivation of the method and present experimental results on polynomial feature extraction for pattern recognition.},
added-at = {2016-04-14T01:01:43.000+0200},
author = {Sch{\"o}lkopf, Bernhard and Smola, Alexander and M{\"u}ller, Klaus-Robert},
biburl = {https://www.bibsonomy.org/bibtex/2c4671c14f5daddd8a25ea1746943368f/peter.ralph},
booktitle = {Artificial Neural Networks-ICANN'97},
interhash = {6473d3f5d5dac85a9d3a2a8cfef57d35},
intrahash = {c4671c14f5daddd8a25ea1746943368f},
keywords = {PCA kernel_PCA linear_algebra methods support_vector_machines},
pages = {583--588},
publisher = {Springer},
timestamp = {2016-04-14T01:01:43.000+0200},
title = {Kernel principal component analysis},
url = {http://link.springer.com/chapter/10.1007/BFb0020217},
year = 1997
}