How to speed up the training mechanism in a connectionist model - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2005

How to speed up the training mechanism in a connectionist model

Szilárd Vajda
  • Function : Author
  • PersonId : 830136
Abdel Belaïd
  • Function : Author
  • PersonId : 830137

Abstract

In this paper a fast data driven learning-corpus building algorithm (FDDLCB) is proposed. The generic technique allows to build dynamically a representative and compact learning corpus for a connectionist model. The constructed dataset contains just a reduced number of patterns but sufficiently descriptive to characterize the different classes which should be separated. The method is based on a double least mean squares (LMS) error minimization mechanism trying to find the optimal boundaries of the different pattern classes. In the classical learning process the LMS is serving to minimize the error during the learning and this process is improved with a second one, as the new samples selection is also based on the idea to minimize the recognition error. Reinforcing the class boundaries where the recognition fails let us achieve a rapid and good generalization without any loss of occuracy. A modified version of the algorithm will be also presented. The experiments were performed on MNIST (Modified NIST) separated digit dataset. The encouraging result (98.51%) using just 1.85% of the available patterns from the original training dataset is comparable even with the state of the art techniques.
Embargoed file
Embargoed file
Visibility date is undetermined
Loading...

Dates and versions

inria-00000358 , version 1 (27-09-2005)

Identifiers

  • HAL Id : inria-00000358 , version 1

Cite

Szilárd Vajda, Abdel Belaïd. How to speed up the training mechanism in a connectionist model. IAPR - TC3 International Workshop on Neural Networks and Learning in Document Recognition - NNLDAR 2005, Aug 2005, Seoul/Korea, pp.13-17. ⟨inria-00000358⟩
49 View
0 Download

Share

Gmail Facebook X LinkedIn More