Risk upper bounds for general ensemble methods with an application to multiclass classification.
François LavioletteEmilie MorvantLiva RalaivolaJean-Francis RoyPublished in: Neurocomputing (2017)
Keyphrases
- ensemble methods
- multiclass classification
- upper bound
- binary classification
- base classifiers
- multi class
- base learners
- random forests
- prediction accuracy
- machine learning methods
- ensemble learning
- benchmark datasets
- lower bound
- decision trees
- random forest
- test data
- machine learning algorithms
- generalization ability
- learning scheme
- cost sensitive
- generalization error
- cross validation
- active learning
- training data