Randomness in Tree Ensemble Methods

Authors

Joran Elias, The University of Montana

Document Type

Presentation Abstract

Presentation Date

4-20-2009

Abstract

In this talk we will discuss classification and regression trees as statistical learners, particularly their use as part of an ensemble. We will discuss the increasing role that randomness has played in developing new tree ensemble techniques. Specifically, it has remained somewhat unclear as to why the simple act of randomizing the tree growing algorithm should lead to the dramatic improvements in performance that have been observed. We will introduce the recent suggestion that a specific type of tree ensemble acts by forming a locally adaptive distance metric as a possible explanation and briefly outline a generalization of this claim to include all tree ensemble methods.

Additional Details

Monday, 20 April 2009
3:10 p.m. in Math 103
4:00 p.m. Refreshments in Math Lounge 109

Recommended Citation

Elias, Joran, "Randomness in Tree Ensemble Methods" (2009). Colloquia of the Department of Mathematical Sciences. 316.
https://scholarworks.umt.edu/mathcolloquia/316