eprintid: 1710
rev_number: 6
eprint_status: archive
userid: 46
dir: disk0/00/00/17/10
datestamp: 2013-09-13 09:40:22
lastmod: 2013-09-16 12:03:00
status_changed: 2013-09-13 09:40:22
type: article
metadata_visibility: no_search
creators_name: Gnecco, Giorgio
creators_name: Sanguineti, Marcello
creators_id: giorgio.gnecco@imtlucca.it
creators_id: 
title: Error bounds for suboptimal solutions to kernel principal component analysis
ispublished: pub
subjects: QA75
divisions: CSA
full_text_status: none
keywords: Principal component analysis (PCA); Kernel methods; Suboptimal solutions
abstract: Suboptimal solutions to kernel principal component analysis are considered. Such solutions take on the form of linear combinations of all n-tuples of kernel functions centered on the data, where n is a positive integer smaller than the cardinality m of the data sample. Their accuracy in approximating the optimal solution, obtained in general for n = m, is estimated. The analysis made in Gnecco and Sanguineti (Comput Optim Appl 42:265–287, 2009) is extended. The estimates derived therein for the approximation of the first principal axis are improved and extensions to the successive principal axes are derived.
date: 2010
date_type: published
publication: Optimization Letters
volume: 4
number: 2
publisher: Springer
pagerange: 197-210
id_number: 10.1007/s11590-009-0158-1
refereed: TRUE
issn: 1862-4472
official_url: http://dx.doi.org/10.1007/s11590-009-0158-1
citation:   Gnecco, Giorgio and Sanguineti, Marcello  Error bounds for suboptimal solutions to kernel principal component analysis.  Optimization Letters, 4 (2).  pp. 197-210.  ISSN 1862-4472      (2010)