Extremely Randomized Trees

This paper proposes a new tree-based ensemble method for supervised classication and regression problems. It essentially consists of randomizing strongly both attribute and cut-point choice while splitting a tree node. In the extreme case, it builds totally randomized trees whose structures are independent of the output values of the learning sample. The strength of the randomization can be tuned to problem specics by the appropriate choice of a parameter. We evaluate the robustness of the default choice of this parameter, and we also provide insight on how to adjust it in particular situations. Besides accuracy, the main strength of the resulting algorithm is computational e ciency. A bias/variance analysis of the Extra-Trees algorithm is also provided as well as a geometrical and a kernel characterization of the models induced.

@Article\{GEW06a,
  author       = "Geurts, Pierre and Ernst, Damien and Wehenkel, Louis",
  title        = "Extremely Randomized Trees",
  journal      = "Machine Learning",
  number       = "1",
  volume       = "36",
  pages        = "3-42",
  year         = "2006",
  keywords     = "machine learning",
  url          = "http://www.montefiore.ulg.ac.be/services/stochastic/pubs/2006/GEW06a"
}