Regularized boosting with an increasing coefficient magnitude stop criterion as meta-learner in hyperparameter optimization stacking ensemble.
Laura Fdez-DíazJosé Ramón QuevedoElena MontañésPublished in: Neurocomputing (2023)
Keyphrases
- ensemble learning
- base classifiers
- ensemble methods
- base learners
- weak learners
- combining classifiers
- risk minimization
- multi class
- feature selection
- combining multiple
- weak classifiers
- optimization problems
- optimization algorithm
- decision trees
- learning algorithm
- neural network
- ensemble classifier
- prediction accuracy
- least squares
- learning environment
- ensemble classification
- optimization criteria
- generalization ability
- optimization method
- random forests
- random forest
- multiple classifiers
- training samples
- weighted voting
- regression problems
- kernel machines
- loss function
- individual classifiers
- semi supervised
- objective function
- e learning
- grid search
- randomized trees