[R] R helps win competitions

Ravi Varadhan rvaradhan at jhmi.edu
Wed Mar 23 17:55:59 CET 2011


Ruben,

Thanks for bringing attention to this very interesting article.  

The Kaggle competition model is aimed at identifying a "single" best
prediction machine.  I am curious as to whether the Kaggle model described
in the article can be extended to an ensemble "uber-learner", where one can
combine the individual prediction models in Kaggle to obtain a more powerful
predictor.  I know that Chris Volinsky and others have used this ensemble
learner idea in winning their Netflix competition.  Is this a practically
feasible idea that can be implemented within the Kaggle system so that at
the end of the competition an uber-learner (that is better than the best
individual prediction) is automatically developed?  Also, would there be
challenges associated with the interpretability and portability of the
resulting uber-learner?  

I would love to hear thoughts from the R prediction experts.

Thanks & Best,
Ravi.

-------------------------------------------------------
Ravi Varadhan, Ph.D.
Assistant Professor,
Division of Geriatric Medicine and Gerontology School of Medicine Johns
Hopkins University

Ph. (410) 502-2619
email: rvaradhan at jhmi.edu


-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On
Behalf Of Rubén Roa
Sent: Wednesday, March 23, 2011 12:10 PM
To: r-help at r-project.org
Subject: [R] R helps win competitions


DeaR ComRades,

This is a quote from a News article in Science's 11-February issue, about
competitions to model data:

"For Chris Raimondi, a search-engine expert based in Baltimore, Maryland,
and winner of the HIV-treatment competition, the Kaggle contest motivated
him to hone his skills in a newly learned computer language called R, which
he used to encode the winning data model. Raimondi also enjoys the
competitive aspect of Kaggle challenges: "It was nice to be able to compare
yourself with others; ... it became kind of addictive. ... I spent more time
on this than I should."

If you are interested read the full article here:

http://www.sciencemag.org/content/331/6018/698.full


Rubén

____________________________________________________________________________
________ 

Dr. Rubén Roa-Ureta
AZTI - Tecnalia / Marine Research Unit
Txatxarramendi Ugartea z/g
48395 Sukarrieta (Bizkaia)
SPAIN

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.



More information about the R-help mailing list