File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Sparse-MIML: A sparsity-based multi-instance multi-learning algorithm

TitleSparse-MIML: A sparsity-based multi-instance multi-learning algorithm
Authors
KeywordsSparsity
iterative methods
label ranking
Markov chain
multi-instance multi-label data
Issue Date2013
PublisherSpringer.
Citation
9th International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR 2013), Lund, Sweden, 19-21 August 2013. In Energy Minimization Methods in Computer Vision and Pattern Recognition: 9th International Conference, EMMCVPR 2013, Lund, Sweden, August 19-21, 2013: Proceedings, 2013, p. 294-306 How to Cite?
AbstractMulti-Instance Multi-Label (MIML) learning is one of challenging research problems in machine learning. The main aim of this paper is to propose and develop a novel sparsity-based MIML learning algorithm. Our idea is to formulate and construct a transductive objective function for labels indicator to be learned by using the method of random walk with restart that exploits the relationships among instances and labels of objects, and computes the affinities among the objects. Then sparsity can be introduced in the labels indicator of the objective function such that relevant and irrelevant objects with respect to a given class can be distinguished. The resulting sparsity-based MIML model can be given as a constrained convex optimization problem, and it can be solved very efficiently by using the augmented Lagrangian method. Experimental results on benchmark data have shown that the proposed sparse-MIML algorithm is computationally efficient, and effective in label prediction for MIML data. We demonstrate that the performance of the proposed method is better than the other testing MIML learning algorithms. © 2013 Springer-Verlag.
Persistent Identifierhttp://hdl.handle.net/10722/276962
ISBN
ISSN
2005 Impact Factor: 0.302
2020 SCImago Journal Rankings: 0.249
Series/Report no.Lecture Notes in Computer Science ; 8081

 

DC FieldValueLanguage
dc.contributor.authorShen, Chenyang-
dc.contributor.authorJing, Liping-
dc.contributor.authorNg, Michael K.-
dc.date.accessioned2019-09-18T08:35:11Z-
dc.date.available2019-09-18T08:35:11Z-
dc.date.issued2013-
dc.identifier.citation9th International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR 2013), Lund, Sweden, 19-21 August 2013. In Energy Minimization Methods in Computer Vision and Pattern Recognition: 9th International Conference, EMMCVPR 2013, Lund, Sweden, August 19-21, 2013: Proceedings, 2013, p. 294-306-
dc.identifier.isbn9783642403941-
dc.identifier.issn0302-9743-
dc.identifier.urihttp://hdl.handle.net/10722/276962-
dc.description.abstractMulti-Instance Multi-Label (MIML) learning is one of challenging research problems in machine learning. The main aim of this paper is to propose and develop a novel sparsity-based MIML learning algorithm. Our idea is to formulate and construct a transductive objective function for labels indicator to be learned by using the method of random walk with restart that exploits the relationships among instances and labels of objects, and computes the affinities among the objects. Then sparsity can be introduced in the labels indicator of the objective function such that relevant and irrelevant objects with respect to a given class can be distinguished. The resulting sparsity-based MIML model can be given as a constrained convex optimization problem, and it can be solved very efficiently by using the augmented Lagrangian method. Experimental results on benchmark data have shown that the proposed sparse-MIML algorithm is computationally efficient, and effective in label prediction for MIML data. We demonstrate that the performance of the proposed method is better than the other testing MIML learning algorithms. © 2013 Springer-Verlag.-
dc.languageeng-
dc.publisherSpringer.-
dc.relation.ispartofEnergy Minimization Methods in Computer Vision and Pattern Recognition: 9th International Conference, EMMCVPR 2013, Lund, Sweden, August 19-21, 2013: Proceedings-
dc.relation.ispartofseriesLecture Notes in Computer Science ; 8081-
dc.subjectSparsity-
dc.subjectiterative methods-
dc.subjectlabel ranking-
dc.subjectMarkov chain-
dc.subjectmulti-instance multi-label data-
dc.titleSparse-MIML: A sparsity-based multi-instance multi-learning algorithm-
dc.typeConference_Paper-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-642-40395-8_22-
dc.identifier.scopuseid_2-s2.0-84884945309-
dc.identifier.spage294-
dc.identifier.epage306-
dc.identifier.eissn1611-3349-
dc.publisher.placeBerlin-
dc.identifier.issnl0302-9743-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats