An ensemble of degraded neural networks

Authors:
Eduardo Vázquez-Santacruz;Debrup Chakraborty
Affiliations:
Department of Electrical Engineering and Computer Science, CINVESTAV-IPN, Jalisco and Department of Computer Science, CINVESTAV-IPN, Mexico City, Mexico;Department of Electrical Engineering and Computer Science, CINVESTAV-IPN, Jalisco and Department of Computer Science, CINVESTAV-IPN, Mexico City, Mexico
Venue:
MCPR'11 Proceedings of the Third Mexican conference on Pattern recognition
Year:
2011

Citing 12
Cited 0

Bagging predictors

Machine Learning
A penalty-function approach for pruning feedforward neural networks

Neural Computation
Ensembling neural networks: many could be better than all

Artificial Intelligence
Neural Network Ensembles

IEEE Transactions on Pattern Analysis and Machine Intelligence
Measures of Diversity in Classifier Ensembles and Their Relationship with the Ensemble Accuracy

Machine Learning
Improving Performance in Neural Networks Using a Boosting Algorithm

Advances in Neural Information Processing Systems 5, [NIPS Conference]
Theoretical Views of Boosting

EuroCOLT '99 Proceedings of the 4th European Conference on Computational Learning Theory
Classifier Ensembles with a Random Linear Oracle

IEEE Transactions on Knowledge and Data Engineering
An Improved Bagging Neural Network Ensemble Algorithm and Its Application

ICNC '07 Proceedings of the Third International Conference on Natural Computation - Volume 05
Novel Approaches to Probabilistic Neural Networks Through Bagging and Evolutionary Estimating of Prior Probabilities

Neural Processing Letters
A brief introduction to boosting

IJCAI'99 Proceedings of the 16th international joint conference on Artificial intelligence - Volume 2
Asymptotic statistical theory of overtraining and cross-validation

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present a new method to create neural network ensembles. In an ensemble method like bagging one needs to train multiple neural networks to create the ensemble. Here we present a scheme to generate different copies of a network from one trained network, and use those copies to create the ensemble. The copies are produced by adding controlled noise to a trained base network. We provide a preliminary theoretical justification for our method and experimentally validate the method on several standard data sets. Our method can improve the accuracy of a base network and give rise to considerable savings in training time compared to bagging.