Population theory for boosting ensembles
Breiman, Leo
Ann. Statist., Tome 32 (2004) no. 1, p. 1-11 / Harvested from Project Euclid
Text on Project Euclid
PDF
Alt PDF
Tree ensembles are looked at in distribution space, that is, the limit case of "infinite" sample size. It is shown that the simplest kind of trees is complete in D-dimensional $L_2(P)$ space if the number of terminal nodes T is greater than D. For such trees we show that the AdaBoost algorithm gives an ensemble converging to the Bayes risk.
Publié le : 2004-02-14
Classification:  Trees,  AdaBoost,  Bayes risk,  62H30,  68T10,  68T05 
@article{1079120126,
     author = {Breiman, Leo},
     title = {Population theory for boosting ensembles},
     journal = {Ann. Statist.},
     volume = {32},
     number = {1},
     year = {2004},
     pages = { 1-11},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1079120126}
}

Breiman, Leo. Population theory for boosting ensembles. Ann. Statist., Tome 32 (2004) no. 1, pp.  1-11. http://gdmltest.u-ga.fr/item/1079120126/