Learning to rank using privileged information
Author(s)
Sharmanska, Viktoriia
Quadrianto, Novi
Lampert, Christoph H
Type
Conference Paper
Abstract
Many computer vision problems have an asymmetric distribution of information between training and test time. In this work, we study the case where we are given additional information about the training data, which however will not be available at test time. This situation is called learning using privileged information (LUPI). We introduce two maximum-margin techniques that are able to make use of this additional source of information, and we show that the framework is applicable to several scenarios that have been studied in computer vision before. Experiments with attributes, bounding boxes, image tags and rationales as additional information in object classification show promising results.
Date Issued
2014-03-03
Date Acceptance
2013-09-02
Citation
2013 IEEE International Conference on Computer Vision, 2014, pp.825-832
ISBN
978-1-4799-2839-2
ISSN
1550-5499
Publisher
IEEE
Start Page
825
End Page
832
Journal / Book Title
2013 IEEE International Conference on Computer Vision
Copyright Statement
© 2013 IEEE. All rights reserved. This research paper is the Open Access version, provided by the Computer Vision Foundation.
Identifier
http://openaccess.thecvf.com/content_iccv_2013/papers/Sharmanska_Learning_to_Rank_2013_ICCV_paper.pdf
Source
IEEE International Conference on Computer Vision (ICCV)
Subjects
Computer Science, Artificial Intelligence
Computer Science
Machine Learning
Pattern Recognition, Automated
Publication Status
Published
Start Date
2013-12-01
Finish Date
2013-12-08
Coverage Spatial
Sydney, Australia
Date Publish Online
2013-12-01