Stochastic learning of multi-instance dictionary for earth mover’s distance-based histogram comparison

Handle URI:
http://hdl.handle.net/10754/622254
Title:
Stochastic learning of multi-instance dictionary for earth mover’s distance-based histogram comparison
Authors:
Fan, Jihong; Liang, Ru-Ze
Abstract:
Dictionary plays an important role in multi-instance data representation. It maps bags of instances to histograms. Earth mover’s distance (EMD) is the most effective histogram distance metric for the application of multi-instance retrieval. However, up to now, there is no existing multi-instance dictionary learning methods designed for EMD-based histogram comparison. To fill this gap, we develop the first EMD-optimal dictionary learning method using stochastic optimization method. In the stochastic learning framework, we have one triplet of bags, including one basic bag, one positive bag, and one negative bag. These bags are mapped to histograms using a multi-instance dictionary. We argue that the EMD between the basic histogram and the positive histogram should be smaller than that between the basic histogram and the negative histogram. Base on this condition, we design a hinge loss. By minimizing this hinge loss and some regularization terms of the dictionary, we update the dictionary instances. The experiments over multi-instance retrieval applications shows its effectiveness when compared to other dictionary learning methods over the problems of medical image retrieval and natural language relation classification. © 2016 The Natural Computing Applications Forum
KAUST Department:
King Abdullah University of Science and Technology, Thuwal, Saudi Arabia
Citation:
Fan J, Liang R-Z (2016) Stochastic learning of multi-instance dictionary for earth mover’s distance-based histogram comparison. Neural Computing and Applications. Available: http://dx.doi.org/10.1007/s00521-016-2603-2.
Publisher:
Springer Nature
Journal:
Neural Computing and Applications
Issue Date:
17-Sep-2016
DOI:
10.1007/s00521-016-2603-2
Type:
Article
ISSN:
0941-0643; 1433-3058
Sponsors:
The work was funded by Science and Technology project under Grant No. 12531826 of Education Department, Heilongjiang, China.
Additional Links:
http://link.springer.com/article/10.1007%2Fs00521-016-2603-2
Appears in Collections:
Articles

Full metadata record

DC FieldValue Language
dc.contributor.authorFan, Jihongen
dc.contributor.authorLiang, Ru-Zeen
dc.date.accessioned2017-01-02T08:42:40Z-
dc.date.available2017-01-02T08:42:40Z-
dc.date.issued2016-09-17en
dc.identifier.citationFan J, Liang R-Z (2016) Stochastic learning of multi-instance dictionary for earth mover’s distance-based histogram comparison. Neural Computing and Applications. Available: http://dx.doi.org/10.1007/s00521-016-2603-2.en
dc.identifier.issn0941-0643en
dc.identifier.issn1433-3058en
dc.identifier.doi10.1007/s00521-016-2603-2en
dc.identifier.urihttp://hdl.handle.net/10754/622254-
dc.description.abstractDictionary plays an important role in multi-instance data representation. It maps bags of instances to histograms. Earth mover’s distance (EMD) is the most effective histogram distance metric for the application of multi-instance retrieval. However, up to now, there is no existing multi-instance dictionary learning methods designed for EMD-based histogram comparison. To fill this gap, we develop the first EMD-optimal dictionary learning method using stochastic optimization method. In the stochastic learning framework, we have one triplet of bags, including one basic bag, one positive bag, and one negative bag. These bags are mapped to histograms using a multi-instance dictionary. We argue that the EMD between the basic histogram and the positive histogram should be smaller than that between the basic histogram and the negative histogram. Base on this condition, we design a hinge loss. By minimizing this hinge loss and some regularization terms of the dictionary, we update the dictionary instances. The experiments over multi-instance retrieval applications shows its effectiveness when compared to other dictionary learning methods over the problems of medical image retrieval and natural language relation classification. © 2016 The Natural Computing Applications Forumen
dc.description.sponsorshipThe work was funded by Science and Technology project under Grant No. 12531826 of Education Department, Heilongjiang, China.en
dc.publisherSpringer Natureen
dc.relation.urlhttp://link.springer.com/article/10.1007%2Fs00521-016-2603-2en
dc.subjectEarth mover’s distanceen
dc.subjectHistogram comparisionen
dc.subjectMedical image retrievalen
dc.subjectMulti-instance dictionaryen
dc.subjectMulti-instance learningen
dc.subjectStochastic learningen
dc.titleStochastic learning of multi-instance dictionary for earth mover’s distance-based histogram comparisonen
dc.typeArticleen
dc.contributor.departmentKing Abdullah University of Science and Technology, Thuwal, Saudi Arabiaen
dc.identifier.journalNeural Computing and Applicationsen
dc.contributor.institutionQiqihar Medical University, Qiqihar, Heilongjiang, Chinaen
kaust.authorLiang, Ru-Zeen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.