An empirical bias–variance analysis of DECORATE ensemble method at different training sample sizes

DOI10.1080/02664763.2011.620949zbMath1478.68304OpenAlexW2067035760MaRDI QIDQ5126995

Guan-Wei Wang, Chun-Xia Zhang, Jiang-She Zhang

Publication date: 21 October 2020

Published in: Journal of Applied Statistics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1080/02664763.2011.620949

zbMATH Keywords

AdaBoosttraining sample sizerandom forestbias-variance decompositionclassifier combination method

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Nonparametric statistical resampling methods (62G09) Learning and adaptive systems in artificial intelligence (68T05)

Related Items (1)

RandGA: injecting randomness into parallel genetic algorithm for variable selection

Uses Software

• UCI-ml
• AdaBoost.MH
• boost
• SABoost

Cites Work

• Unnamed Item
• Bagging predictors
• Is combining classifiers with stacking better than selecting the best one?
• Taxonomy for characterizing ensemble methods in classification tasks: a review and annotated bibliography
• A stochastic approximation view of boosting
• A local boosting algorithm for solving classification problems
• Simultaneous statistical inference. 2nd ed
• A decision-theoretic generalization of on-line learning and an application to boosting
• Variance and bias for general loss functions
• Inference for the generalization error
• Ensembling neural networks: Many could be better than all
• Arcing classifiers. (With discussion)
• Combining Pattern Classifiers
• A comparison of prediction accuracy, complexity, and training time of thirty-three old and new classification algorithms
• Random forests

This page was built for publication: An empirical bias–variance analysis of DECORATE ensemble method at different training sample sizes