Initialization framework for latent variable models
2014 (English)In: ICPRAM 2014 - Proceedings of the 3rd International Conference on Pattern Recognition Applications and Methods, 2014, 227-232 p.Conference paper (Refereed)
In this paper, we discuss the properties of a class of latent variable models that assumes each labeled sample is associated with set of different features, with no prior knowledge of which feature is the most relevant feature to be used. Deformable-Part Models (DPM) can be seen as good example of such models. While Latent SVM framework (LSVM) has proven to be an efficient tool for solving these models, we will argue that the solution found by this tool is very sensitive to the initialization. To decrease this dependency, we propose a novel clustering procedure, for these problems, to find cluster centers that are shared by several sample sets while ignoring the rest of the cluster centers. As we will show, these cluster centers will provide a robust initialization for the LSVM framework.
Place, publisher, year, edition, pages
2014. 227-232 p.
Classification, Clustering, Latent variable models, Localization, Classification (of information), Software engineering, Cluster centers, Novel clustering, Prior knowledge, Relevant features, Sample sets, Pattern recognition
Computer and Information Science
IdentifiersURN: urn:nbn:se:kth:diva-168883ScopusID: 2-s2.0-84902354502ISBN: 9789897580185OAI: oai:DiVA.org:kth-168883DiVA: diva2:819337
3rd International Conference on Pattern Recognition Applications and Methods, ICPRAM 2014, 6 March 2014 through 8 March 2014, Angers, Loire Valley
QC 201506102015-06-102015-06-092015-06-10Bibliographically approved