An Efficient Partition of Training Data Set Improves Speed andAccuracy of Cascade-correlation Algorithm

Authors:
Igor V. Tetko;Alessandro E. P. Villa
Affiliations:
Laboratoire de Neuro-heuristique, Institut de Physiologie, Université de Lausanne, Rue du Bugnon 7, Lausanne, CH–1005, Switzerland;Laboratoire de Neuro-heuristique, Institut de Physiologie, Université de Lausanne, Rue du Bugnon 7, Lausanne, CH–1005, Switzerland
Venue:
Neural Processing Letters
Year:
1997

Citing 5
Cited 1

The cascade-correlation learning architecture

Advances in neural information processing systems 2
Bagging predictors

Machine Learning
An Enhancement of Generalization Ability in Cascade CorrelationAlgorithm by Avoidance of Overfitting/Overtraining Problem

Neural Processing Letters
Efficient partition of learning data sets for neural network training

Neural Networks
A decision-theoretic generalization of on-line learning and an application to boosting

EuroCOLT '95 Proceedings of the Second European Conference on Computational Learning Theory

Associative Neural Network

Neural Processing Letters

Quantified Score

Hi-index	0.00

Visualization

Abstract

This study extends an application of efficientpartition algorithm (EPA) for artificial neuralnetwork ensemble trained according to CascadeCorrelation Algorithm. We show that EPA allows todecrease the number of cases in learning and validateddata sets. The predictive ability of the ensemblecalculated using the whole data set is not affectedand in some cases it is even improved. It is shownthat a distribution of cases selected by this methodis proportional to the second derivative of theanalyzed function.