AdaBoost is Consistent

Authors:
Peter L. Bartlett;Mikhail Traskin
Affiliations:
-;-
Venue:
The Journal of Machine Learning Research
Year:
2007

Citing 0
Cited 10

Random classification noise defeats all convex potential boosters

Proceedings of the 25th international conference on Machine learning
Collective-agreement-based pruning of ensembles

Computational Statistics & Data Analysis
Boosting with pairwise constraints

Neurocomputing
Deformation of log-likelihood loss function for multiclass boosting

Neural Networks
Hierarchical Estimator

Expert Systems with Applications: An International Journal
New specifics for a hierarchial estimator meta-algorithm

ICAISC'12 Proceedings of the 11th international conference on Artificial Intelligence and Soft Computing - Volume Part II
Boosting k-NN for Categorization of Natural Scenes

International Journal of Computer Vision
A theory of multiclass boosting

The Journal of Machine Learning Research
On the doubt about margin explanation of boosting

Artificial Intelligence
The rate of convergence of AdaBoost

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

The risk, or probability of error, of the classifier produced by the AdaBoost algorithm is investigated. In particular, we consider the stopping strategy to be used in AdaBoost to achieve universal consistency. We show that provided AdaBoost is stopped after n1-ε iterations---for sample size n and ε ∈ (0,1)---the sequence of risks of the classifiers it produces approaches the Bayes risk.