Learning From Very-Few Labeled Examples with Soft Labels

UTSePress Research/Manakin Repository

Search UTSePress Research

Advanced Search


My Account

Show simple item record

dc.contributor.author Mu, Yadong en_US
dc.contributor.author Xu, Min en_US
dc.contributor.author Yan, Shuicheng en_US
dc.contributor.editor Technical Committee en_US
dc.date.accessioned 2012-02-02T11:08:06Z
dc.date.available 2012-02-02T11:08:06Z
dc.date.issued 2010 en_US
dc.identifier 2009007894 en_US
dc.identifier.citation Mu Yadong, Xu Min, and Yan Shuicheng 2010, 'Learning From Very-Few Labeled Examples with Soft Labels', , IEEE Computer Society, Hongkong, , pp. 3869-3872. en_US
dc.identifier.issn 978-1-4244-7993-1 en_US
dc.identifier.other E1 en_US
dc.identifier.uri http://hdl.handle.net/10453/16273
dc.description.abstract In this paper we propose Softboost, a novel Boosting al-gorithm which combines the merits of transductive and inductive learning approaches to attack the problem of learning from very few labeled training examples. In the transductive stage, soft labels of both the labeled and unlabeled samples are estimated based on a Markovian propagating procedure. While in the subsequent inductive stage, to efficiently handle out-of-sample data, we learn a weighted combination of simple rules in Boosting style, each of which maximizes confidence-weighted inter-class Kullback-Leibler (KL) divergence under current data distribution. Finally, experiments on toy dataset and USPS handwritten digits are presented to demonstrate its effectiveness. en_US
dc.language English en_US
dc.publisher IEEE Computer Society en_US
dc.relation.isbasedon NA en_US
dc.title Learning From Very-Few Labeled Examples with Soft Labels en_US
dc.parent 2010 IEEE International Conference on Image Processing ICIP 2010 - Proceedings en_US
dc.journal.volume en_US
dc.journal.number en_US
dc.publocation Hongkong en_US
dc.identifier.startpage 3869 en_US
dc.identifier.endpage 3872 en_US
dc.cauo.name FEIT.School of Computing and Communications en_US
dc.conference Verified OK en_US
dc.for 080106 en_US
dc.personcode 0000064301 en_US
dc.personcode 109684 en_US
dc.personcode 0000064303 en_US
dc.percentage 100 en_US
dc.classification.name Image Processing en_US
dc.classification.type FOR-08 en_US
dc.edition en_US
dc.custom IEEE International Conference on Image Processing en_US
dc.date.activity 20100926 en_US
dc.location.activity Hongkong en_US
dc.description.keywords Boosting, soft label en_US

Files in this item

This item appears in the following Collection(s)

Show simple item record