[Statlist] Talk on Friday, September 11, 2009

[rey m@iii@g oii st@t@m@th@ethz@ch]

R E M I N D E R

_________________________________________________
ETH and University of Zurich

Proff. A.D. Barbour - P. Buehlmann -  L. Held -
H.R. Kuensch - M. Maathuis - W. Stahel - S. van de Geer


*********************************************************
We are glad to announce the following talk

*Friday, September 11, 2009 15.15 - 17.00  HG  G 19.1 *

***********************************************************
*with Nicolai Meinshausen, University of Oxford, Oxford UK*

Title:

Trees, Forests and Group aggregation


Abstract:

In higher-dimensional regression and classification, there is a natural
tradeoff between simplicity of the algorithm and its predictive power.
Simple procedure like trees are intuitive to understand, yet they are
clearly beaten in terms of predictive accuracy by more complex methods like
tree ensembles, including Random Forests.
I will share some thoughts and notes on this and show that a convex
relaxation
to an optimal partitioning of the data yields a new algorithm,
which I call group aggregation.
Predictions are simply weighted averages over empirical group means,
for suitably selected groups of observations.
The group selection amounts to a quadratic programming problem and can
be solved efficiently.
Even though the algorithm contains no explicit tuning parameters in
the most simple version,
group aggregation gets close to Random Forests predictive power,
while maintaining or surpassing the simplicity of trees.
An application to emulation in climate modeling will be discussed.



This abstract is also to be found under the following link:
http://stat.ethz.ch/talks/research_seminar

-- 
ETH Zürich
Seminar für Statistik
Cecilia Rey-Lutz, HG G10.3
Rämistrasse 101
CH-8092 Zurich
mail: rey using stat.math.ethz.ch
phone: +41 44 632 3438/fax: +41 44 632 1228

_______________________________________________
Statlist mailing list
Statlist using stat.ch
https://stat.ethz.ch/mailman/listinfo/statlist