Resilient approximation of kernel classifiers

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

Trained support vector machines (SVMs) have a slow run-time classification speed if the classification problem is noisy and the sample data set is large. Approximating the SVM by a more sparse function has been proposed to solve to this problem. In this study, different variants of approximation algorithms are empirically compared. It is shown that gradient descent using the improved Rprop algorithm increases the robustness of the method compared to fixed-point iteration. Three different heuristics for selecting the support vectors to be used in the construction of the sparse approximation are proposed. It turns out that none is superior to random selection. The effect of a finishing gradient descent on all parameters of the sparse approximation is studied.

OriginalsprogEngelsk
TitelArtificial Neural Networks – ICANN 2007 : 17th International Conference, Porto, Portugal, September 9-13, 2007, Proceedings, Part I
RedaktørerJoaquim Marques de Sá, Lius A. Alexandre, Włodzisław Duch, Danilo Mandic
Antal sider10
Vol/bindPart I
ForlagSpringer
Publikationsdato2007
Sider139-148
ISBN (Trykt)978-3-540-74689-8
ISBN (Elektronisk)978-3-540-74690-4
DOI
StatusUdgivet - 2007
Eksternt udgivetJa
Begivenhed17th International Conference on Artificial Neural Networks, ICANN 2007 - Porto, Portugal
Varighed: 9 sep. 200713 sep. 2007

Konference

Konference17th International Conference on Artificial Neural Networks, ICANN 2007
LandPortugal
ByPorto
Periode09/09/200713/09/2007
NavnLecture notes in computer science
Vol/bind4668
ISSN0302-9743

ID: 168563567