crossvalidationnetwork parameters and architecture optimizationneural network ensemblesresidual generalization error
Figures from the paper
Abstract (AI)
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.>
Key Findings
1
Cross-validation can be used to optimize neural network parameters and architecture for classification tasks.
2
Ensembles of similar neural networks reduce the remaining residual generalization error.
3
Using multiple means (including cross-validation and ensembles) improves performance and training of classification neural networks.
Research Object
Neural network classifiers (ensembles of similar neural networks)
Research Subject
Methods to improve performance and training, including cross-validation for parameter/architecture selection and reduction of residual generalization error via ensembles
Publication Details
Publication Date
1990-01-01
Journal
Publisher
ISSN
Cited by
4257
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
Cited by7
Random decision forests2002
Neural Networks and the Bias/Variance Dilemma1992
A survey of uncertainty in deep neural networks2023
Efficient Deep Learning: A Survey on Making Deep Learning Models Smaller, Faster, and Better2023
An Intelligent Time-Adaptive Data-Driven Method for Sensor Fault Diagnosis in Induction Motor Drive System2018
Unplanned dilution and ore loss prediction in longhole stoping mines via multiple regression and artificial neural network analyses2015
Artificial Intelligence and Parametric Construction Cost Estimate Modeling: State-of-the-Art Review2019