Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/67027
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Journal article |
Title: | UBoost: Boosting with the Universum |
Other Titles: | {cal U}Boost: Boosting with the Universum |
Author: | Shen, C. Wang, P. Shen, F. Wang, H. |
Citation: | IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012; 34(4):825-832 |
Publisher: | IEEE Computer Soc |
Issue Date: | 2012 |
ISSN: | 0162-8828 1939-3539 |
Statement of Responsibility: | Chunhua Shen, Peng Wang, Fumin Shen and Hanzi Wang |
Abstract: | It has been shown that the Universum data, which do not belong to either class of the classification problem of interest, may contain useful prior domain knowledge for training a classifier [1], [2]. In this work, we design a novel boosting algorithm that takes advantage of the available Universum data, hence the name UBoost. UBoost is a boosting implementation of Vapnik’s alternative capacity concept to the large margin approach. In addition to the standard regularization term, UBoost also controls the learned model’s capacity by maximizing the number of observed contradictions. Our experiments demonstrate that UBoost can deliver improved classification accuracy over standard boosting algorithms that use labeled data alone. |
Keywords: | Universum kernel methods boosting column generation convex optimization |
Rights: | Copyright 2012 IEEE |
DOI: | 10.1109/TPAMI.2011.240 |
Published version: | http://dx.doi.org/10.1109/tpami.2011.240 |
Appears in Collections: | Aurora harvest Computer Science publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
hdl_67027.pdf | Accepted version | 481.95 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.