Please use this identifier to cite or link to this item: https://hdl.handle.net/2440/70663
Citations
Scopus Web of Science® Altmetric
?
?
Full metadata record
DC FieldValueLanguage
dc.contributor.authorWang, T.-
dc.contributor.authorHe, X.-
dc.contributor.authorShen, C.-
dc.contributor.authorBarnes, N.-
dc.date.issued2011-
dc.identifier.citationProceedings of International Conference on on Digital Image Computing: Techniques and Applications: DICTA 2011: pp.209-216-
dc.identifier.isbn9780769545882-
dc.identifier.urihttp://hdl.handle.net/2440/70663-
dc.description.abstractBoosting algorithms attract much attention in computer vision and image processing because of their strong performance in a variety of applications. Recent progress on the theory of boosting algorithms suggests a close link between good generalization and the margin distribution of the classifier w.r.t. a dataset. In this paper, we propose a novel data-dependent margin distribution learning criterion for boosting, termed Laplacian MDBoost, which utilizes the intrinsic geometric structure of dataset. One key aspect of our method is that it can seamlessly incorporate unlabeled data by including a graph Laplacian regularizer. We derive a dual formulation of the learning problem that can be efficiently solved by column generation. Experiments on various datasets validate the effectiveness of the new graph Laplacian based learning criterion on both supervised and unsupervised learning settings. We also show that the performance of our algorithm outperforms the state-of-the-art semi-supervised learning algorithms on a variety of inductive inference tasks, including real world video segmentation.-
dc.description.statementofresponsibilityTao Wang, Xuming He, Chunhua Shen, and Nick Barnes-
dc.description.urihttp://itee.uq.edu.au/~dicta2011/-
dc.language.isoen-
dc.publisherIEEE-
dc.rightsCopyright © 2011 by The Institute of Electrical and Electronics Engineers, Inc.-
dc.source.urihttp://dx.doi.org/10.1109/dicta.2011.42-
dc.subjectmargin distribution boosting-
dc.subjectLaplacian eigenmaps-
dc.subjectsemi-supervised learning-
dc.titleLaplacian margin distribution boosting for learning from sparsely labeled data-
dc.typeConference paper-
dc.contributor.conferenceDigital Image Computing Techniques and Applications (2011 : Noosa, Qld.)-
dc.identifier.doi10.1109/DICTA.2011.42-
dc.publisher.placeUSA-
pubs.publication-statusPublished-
dc.identifier.orcidShen, C. [0000-0002-8648-8718]-
Appears in Collections:Aurora harvest 5
Computer Science publications

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.