160 likes | 454 Views
Popular Ensemble Methods: An Empirical Study. David Opitz and Richard Maclin Presented by Scott Wespi 5/22/07. Outline. Ensemble methods Classifier Ensembles Bagging vs Boosting Results Conclusion. Ensemble Methods.
E N D
Popular Ensemble Methods: An Empirical Study David Opitz and Richard Maclin Presented by Scott Wespi 5/22/07
Outline • Ensemble methods • Classifier Ensembles • Bagging vs Boosting • Results • Conclusion
Ensemble Methods • Sets of individually trained classifiers whose predictions are combined when classifying new data • Bagging (1996) • Boosting (1996) • How are bagging and boosting influenced by the learning algorithm? • Decision trees • Neural networks *Note: Paper is from 1999
Classifier Ensembles • Goal: highly accurate individual classifiers that disagree as much as possible • Bagging and boosting create disagreement
Ada-Boosting vs Arcing • Ada-Boosting • Every sample has 1/N weight initially, increases every time sample was skipped or misclassified • Arcing • If mi = number of times ith example was misclassified
Neural Networks • Ada-Boosting • Arcing • Bagging • White bar represents 1 • standard deviation
Neural Networks: Bagging vs Simple
Ada-Boost: Neural Networks vs. Decision Trees • NN • DT • Box represents • reduction in error
Noise • Hurts boosting the most
Conclusions • Performance depends on data and classifier • In some cases, ensembles can overcome bias of component learning algorithm • Bagging is more consistent than boosting • Boosting can give much better results on some data