Title :
Neural network ensembles
Author :
Hansen, Lars Kai ; Salamon, Peter
Author_Institution :
Dept. of Math. Sci., San Diego State Univ., CA, USA
fDate :
10/1/1990 12:00:00 AM
Abstract :
Several means for improving the performance and training of neural networks for classification are proposed. Crossvalidation is used as a tool for optimizing network parameters and architecture. It is shown that the remaining residual generalization error can be reduced by invoking ensembles of similar networks
Keywords :
learning systems; neural nets; pattern recognition; classification; crossvalidation; neural networks; pattern recognition; performance; residual generalization error; training; Computer architecture; Data mining; Databases; Fault tolerance; Feedforward systems; Neural networks; Neurons; Pattern recognition; Performance analysis; Supervised learning;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
