Show simple item record

dc.contributor.authorWang, Jim Jing-Yan
dc.contributor.authorTsang, Ivor Wai-Hung
dc.contributor.authorCui, Xuefeng
dc.contributor.authorLu, Zhiwu
dc.contributor.authorGao, Xin
dc.date.accessioned2017-01-02T13:44:59Z
dc.date.available2017-01-02T13:44:59Z
dc.date.issued2016-12-29
dc.identifier.citationWang JJ-Y, Tsang IW-H, Cui X, Lu Z, Gao X (2016) Multi-instance dictionary learning via multivariate performance measure optimization. Pattern Recognition. Available: http://dx.doi.org/10.1016/j.patcog.2016.12.023.
dc.identifier.issn0031-3203
dc.identifier.doi10.1016/j.patcog.2016.12.023
dc.identifier.urihttp://hdl.handle.net/10754/622623
dc.description.abstractThe multi-instance dictionary plays a critical role in multi-instance data representation. Meanwhile, different multi-instance learning applications are evaluated by specific multivariate performance measures. For example, multi-instance ranking reports the precision and recall. It is not difficult to see that to obtain different optimal performance measures, different dictionaries are needed. This observation motives us to learn performance-optimal dictionaries for this problem. In this paper, we propose a novel joint framework for learning the multi-instance dictionary and the classifier to optimize a given multivariate performance measure, such as the F1 score and precision at rank k. We propose to represent the bags as bag-level features via the bag-instance similarity, and learn a classifier in the bag-level feature space to optimize the given performance measure. We propose to minimize the upper bound of a multivariate loss corresponding to the performance measure, the complexity of the classifier, and the complexity of the dictionary, simultaneously, with regard to both the dictionary and the classifier parameters. In this way, the dictionary learning is regularized by the performance optimization, and a performance-optimal dictionary is obtained. We develop an iterative algorithm to solve this minimization problem efficiently using a cutting-plane algorithm and a coordinate descent method. Experiments on multi-instance benchmark data sets show its advantage over both traditional multi-instance learning and performance optimization methods.
dc.description.sponsorshipWe are grateful to Prof. Zhi-Hua Zhou for the fruitful discussions. The research reported in this publication was supported by funding from King Abdullah University of Science and Technology (KAUST).
dc.publisherElsevier BV
dc.relation.urlhttp://www.sciencedirect.com/science/article/pii/S0031320316304435
dc.rightsNOTICE: this is the author’s version of a work that was accepted for publication in Pattern Recognition. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Pattern Recognition, 29 December 2016. DOI: 10.1016/j.patcog.2016.12.023
dc.subjectMulti-instance learning
dc.subjectDictionary
dc.subjectMultivariate performance measures
dc.subjectCutting-plane algorithm
dc.titleMulti-instance dictionary learning via multivariate performance measure optimization
dc.typeArticle
dc.contributor.departmentComputational Bioscience Research Center (CBRC)
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.identifier.journalPattern Recognition
dc.eprint.versionPost-print
dc.contributor.institutionCentre for Quantum Computation and Intelligent Systems, University of Technology Sydney, Australia
dc.contributor.institutionBeijing Key Laboratory of Big Data Management and Analysis Methods, School of Information, Renmin University of China, Beijing, 100872, China
kaust.personWang, Jim Jing-Yan
kaust.personCui, Xuefeng
kaust.personGao, Xin
dc.date.published-online2016-12-29
dc.date.published-print2017-06


Files in this item

Thumbnail
Name:
1-s2.0-S0031320316304435-main.pdf
Size:
610.0Kb
Format:
PDF
Description:
Accepted Manuscript

This item appears in the following Collection(s)

Show simple item record