Parallel Perceptrons, Activation Margins and Imbalanced Training Set Pruning
Entity
UAM. Departamento de Ingeniería InformáticaPublisher
Springer Berlin HeidelbergDate
2005Citation
10.1007/11492542_6
Pattern Recognition and Image Analysis: Second Iberian Conference, IbPRIA 2005, Estoril, Portugal, June 7-9, 2005, Proceedings, Part II. Lecture Notes in Computer Science, Volumen 3523. Springer, 2005. 23-50.
ISSN
0302-9743 (print); 1611-3349 (online)ISBN
978-3-540-26154-4 (print); 978-3-540-32238-2 (online)DOI
10.1007/11492542_6Funded by
With partial support of Spain’s CICyT, TIC 01–572, TIN2004–07676Editor's Version
http://dx.doi.org/10.1007/11492542_6Subjects
Computer Vision; Image Processing; Computer Graphics; Pattern Recognition; Informática; TelecomunicacionesNote
The final publication is available at Springer via http://dx.doi.org/10.1007/11492542_6Proceedings of Second Iberian Conference, IbPRIA 2005, Estoril, Portugal, June 7-9, 2005, Part II
Rights
© Springer-Verlag Berlin Heidelberg 2005Abstract
A natural way to deal with training samples in imbalanced class problems is to prune them removing redundant patterns, easy to classify and probably over represented, and label noisy patterns that belonging to one class are labelled as members of another. This allows classifier construction to focus on borderline patterns, likely to be the most informative ones. To appropriately define the above subsets, in this work we will use as base classifiers the so–called parallel perceptrons, a novel approach to committee machine training that allows, among other things, to naturally define margins for hidden unit activations. We shall use these margins to define the above pattern types and to iteratively perform subsample selections in an initial training set that enhance classification accuracy and allow for a balanced classifier performance even when class sizes are greatly different.
Files in this item
Google Scholar:Cantador Gutiérrez, Iván
-
Dorronsoro Ibero, José Ramón
This item appears in the following Collection(s)
Related items
Showing items related by title, author, creator and subject.