A Fast Learning Strategy Using Pattern Selection for Feedforward Neural Networks

Abstract : Intelligent pattern selection is an active learning strategy where the classifiers select during training the most informative patterns. This paper investigates such a strategy where the informativeness of a pattern is measured as the approximation error produced by the classifier. The algorithm builds the training corpus starting from a small randomly chosen initial dataset and new patterns are added to the learning corpus based on their error sensitivity. The training dataset expansion is based on the selection of the most erroneous patterns. Our experimental results on MNIST 1 separated digit dataset show that only 3.26%of training data are sufficient for training purpose without decreasing the performance (98.36%) of the resulting neural classifier.
Document type :
Conference papers
Guy Lorette. Tenth International Workshop on Frontiers in Handwriting Recognition 2006 - IWFHR'10, Oct 2006, La Baule, France. Suvisoft, pp.6, 2006


https://hal.inria.fr/inria-00104833
Contributor : Anne Jaigu <>
Submitted on : Monday, October 9, 2006 - 2:59:40 PM
Last modification on : Wednesday, December 8, 2010 - 5:24:19 PM

Identifiers

  • HAL Id : inria-00104833, version 1

Collections

Citation

Szilárd Vajda, Yves Rangoni, Hubert Cecotti, Abdel Belaïd. A Fast Learning Strategy Using Pattern Selection for Feedforward Neural Networks. Guy Lorette. Tenth International Workshop on Frontiers in Handwriting Recognition 2006 - IWFHR'10, Oct 2006, La Baule, France. Suvisoft, pp.6, 2006. <inria-00104833>

Export

Share

Metrics

Consultation de
la notice

90

Téléchargement du document

32