A Caltech Library Service

Quickly Boosting Decision Trees - Pruning Underachieving Features Early

Appel, Ron and Fuchs, Thomas and Dollár, Piotr and Perona, Pietro (2013) Quickly Boosting Decision Trees - Pruning Underachieving Features Early. JMLR Workshop and Conference Proceedings, 28 . pp. 594-602. ISSN 1938-7228.

PDF - Published Version
See Usage Policy.

PDF - Supplemental Material
See Usage Policy.


Use this Persistent URL to link to this item:


Boosted decision trees are one of the most popular and successful learning techniques used today. While exhibiting fast speeds at test time, relatively slow training makes them impractical for applications with real-time learning requirements. We propose a principled approach to overcome this drawback. We prove a bound on the error of a decision stump given its preliminary error on a subset of the training data; the bound may be used to prune unpromising features early on in the training process. We propose a fast training algorithm that exploits this bound, yielding speedups of an order of magnitude at no cost in the final performance of the classifier. Our method is not a new variant of Boosting; rather, it may be used in conjunction with existing Boosting algorithms and other sampling heuristics to achieve even greater speedups.

Item Type:Article
Related URLs:
URLURL TypeDescription
Perona, Pietro0000-0002-7583-5809
Contact Email
Additional Information:Copyright 2013 by the author(s). This work was supported by NSERC 420456-2012, MURI-ONR N00014-10-1-0933, ARO/JPL-NASA Stennis NAS7.03001, and Moore Foundation.
Funding AgencyGrant Number
NSERC (Natural Sciences and Engineering Research Council of Canada)420456-2012
ARO/JPL-NASA StennisNAS7.03001
Gordon and Betty Moore FoundationUNSPECIFIED
Record Number:CaltechAUTHORS:20131007-134123893
Persistent URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:41719
Deposited By: SWORD User
Deposited On:07 Oct 2013 20:54
Last Modified:03 Oct 2019 05:51

Repository Staff Only: item control page