Abstract

Kernel principal component analysis (KPCA) is a method widely used for denoising multivariate data. Using geometric arguments, we investigate why a projection operation inherent to all existing KPCA denoising algorithms can sometimes cause very poor denoising. Based on this, we propose a modification to the projection operation that remedies this problem and can be incorporated into any of the existing KPCA algorithms. Using toy examples and real datasets, we show that the proposed algorithm can substantially improve denoising performance and is more robust to misspecification of an important tuning parameter.

Original languageEnglish (US)
Article number6156791
Pages (from-to)644-656
Number of pages13
JournalIEEE Transactions on Neural Networks and Learning Systems
Volume23
Issue number4
DOIs
StatePublished - 2012

Fingerprint

Principal component analysis
Tuning

Keywords

  • Denoising
  • kernel
  • kernel principal component analysis (KPCA)
  • preimage problem

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Software

Cite this

Tangent hyperplane kernel principal component analysis for denoising. / Im, Joon Ku; Apley, Daniel W.; Runger, George.

In: IEEE Transactions on Neural Networks and Learning Systems, Vol. 23, No. 4, 6156791, 2012, p. 644-656.

Research output: Contribution to journalArticle

@article{e3e0819f48604e5ca86741f85c6131dc,
title = "Tangent hyperplane kernel principal component analysis for denoising",
abstract = "Kernel principal component analysis (KPCA) is a method widely used for denoising multivariate data. Using geometric arguments, we investigate why a projection operation inherent to all existing KPCA denoising algorithms can sometimes cause very poor denoising. Based on this, we propose a modification to the projection operation that remedies this problem and can be incorporated into any of the existing KPCA algorithms. Using toy examples and real datasets, we show that the proposed algorithm can substantially improve denoising performance and is more robust to misspecification of an important tuning parameter.",
keywords = "Denoising, kernel, kernel principal component analysis (KPCA), preimage problem",
author = "Im, {Joon Ku} and Apley, {Daniel W.} and George Runger",
year = "2012",
doi = "10.1109/TNNLS.2012.2185950",
language = "English (US)",
volume = "23",
pages = "644--656",
journal = "IEEE Transactions on Neural Networks and Learning Systems",
issn = "2162-237X",
publisher = "IEEE Computational Intelligence Society",
number = "4",

}

TY - JOUR

T1 - Tangent hyperplane kernel principal component analysis for denoising

AU - Im, Joon Ku

AU - Apley, Daniel W.

AU - Runger, George

PY - 2012

Y1 - 2012

N2 - Kernel principal component analysis (KPCA) is a method widely used for denoising multivariate data. Using geometric arguments, we investigate why a projection operation inherent to all existing KPCA denoising algorithms can sometimes cause very poor denoising. Based on this, we propose a modification to the projection operation that remedies this problem and can be incorporated into any of the existing KPCA algorithms. Using toy examples and real datasets, we show that the proposed algorithm can substantially improve denoising performance and is more robust to misspecification of an important tuning parameter.

AB - Kernel principal component analysis (KPCA) is a method widely used for denoising multivariate data. Using geometric arguments, we investigate why a projection operation inherent to all existing KPCA denoising algorithms can sometimes cause very poor denoising. Based on this, we propose a modification to the projection operation that remedies this problem and can be incorporated into any of the existing KPCA algorithms. Using toy examples and real datasets, we show that the proposed algorithm can substantially improve denoising performance and is more robust to misspecification of an important tuning parameter.

KW - Denoising

KW - kernel

KW - kernel principal component analysis (KPCA)

KW - preimage problem

UR - http://www.scopus.com/inward/record.url?scp=84876936094&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876936094&partnerID=8YFLogxK

U2 - 10.1109/TNNLS.2012.2185950

DO - 10.1109/TNNLS.2012.2185950

M3 - Article

AN - SCOPUS:84876936094

VL - 23

SP - 644

EP - 656

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

IS - 4

M1 - 6156791

ER -