Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
Oriented principal component analysis for large margin classifiers
Autore:
Bermejo, S; Cabestany, J;
Indirizzi:
Univ Politecn Catalunya, Dept Elect Engn, ES-08034 Barcelona, Spain Univ Politecn Catalunya Barcelona Spain ES-08034 -08034 Barcelona, Spain
Titolo Testata:
NEURAL NETWORKS
fascicolo: 10, volume: 14, anno: 2001,
pagine: 1447 - 1461
SICI:
0893-6080(200112)14:10<1447:OPCAFL>2.0.ZU;2-A
Fonte:
ISI
Lingua:
ENG
Soggetto:
VECTOR QUANTIZATION; IMAGE COMPRESSION; CLASSIFICATION; PROJECTION;
Keywords:
large margin classifiers; oriented principal component analysis; co-operative learning; principal component neural networks; learning-to-learn algorithms; feature extraction; online gradient descent; pattern recognition;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
Citazioni:
44
Recensione:
Indirizzi per estratti:
Indirizzo: Bermejo, S Univ Politecn Catalunya, Dept Elect Engn, Jordi Girona 1-3,C4 Bldg, ES-08034 Barcelona, Spain Univ Politecn Catalunya Jordi Girona 1-3,C4 Bldg Barcelona Spain ES-08034
Citazione:
S. Bermejo e J. Cabestany, "Oriented principal component analysis for large margin classifiers", NEURAL NETW, 14(10), 2001, pp. 1447-1461

Abstract

Large margin classifiers (such as MLPs) are designed to assign training samples with high confidence (or margin) to one of the classes. Recent theoretical results of these systems show why the use of regularisation terms andfeature extractor techniques can enhance their generalisation properties. Since the optimal subset of features selected depends on the classificationproblem, but also on the particular classifier with which they are used, global learning algorithms for large margin classifiers that use feature extractor techniques are desired. A direct approach is to optimise a cost function based on the margin error, which also incorporates regularisation terms for controlling capacity. These terms must penalise a classifier with thelargest margin for the problem at hand. Our work shows that the inclusion of a PCA term can be employed for this purpose. Since PCA only achieves an optimal discriminatory projection for some particular distribution of data,the margin of the classifier can then be effectively controlled. We also propose a simple constrained search for the global algorithm in which the feature extractor and the classifier are trained separately. This allows a degree of flexibility for including heuristics that can enhance the search and the performance of the computed solution. Experimental results demonstrate the potential of the proposed method. (C) 2001 Elsevier Science Ltd. All rights reserved.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 04/04/20 alle ore 02:40:23