Multiple instance learning via distance metric optimization

Zhao, H; Cheng, J; Jiang, J; Tao, D

Multiple instance learning via distance metric optimization

Zhao, H Cheng, J Jiang, J Tao, D

Permalink

Publication Type:: Conference Proceeding
Citation:: 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings, 2013, pp. 2617 - 2621
Issue Date:: 2013-12-01

Closed Access

	Filename	Description	Size
	2013003393OK.pdf		671.9 kB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Zhao, H	en_US
dc.contributor.author	Cheng, J	en_US
dc.contributor.author	Jiang, J	en_US
dc.contributor.author	Tao, D https://orcid.org/0000-0001-7225-5449	en_US
dc.date.issued	2013-12-01	en_US
dc.identifier.citation	2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings, 2013, pp. 2617 - 2621	en_US
dc.identifier.isbn	9781479923410	en_US
dc.identifier.uri	http://hdl.handle.net/10453/28032
dc.description.abstract	Multiple Instance Learning (MIL) has been widely applied in practice, such as drug activity prediction, content-based image retrieval. In MIL, a sample, comprised of a set of instances, is called a bag. Labels are assigned to bags instead of instances. The uncertainty of labels on instances makes MIL different from conventional supervised single instance learning (SIL) tasks. Therefore, it is critical to learn an effective mapping to convert an MIL task to an SIL task. In this paper, we present OptMILES by learning the optimal transformation on the bag-to-instance similarity measure, exploring the optimal distance metric between instances, by an alternating minimization training procedure. We thoroughly evaluate the proposed method on both a synthetic dataset and real world datasets by comparing with representative MIL algorithms. The experimental results suggest the effectiveness of OptMILES. © 2013 IEEE.	en_US
dc.relation.ispartof	2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings	en_US
dc.relation.isbasedon	10.1109/ICIP.2013.6738539	en_US
dc.title	Multiple instance learning via distance metric optimization	en_US
dc.type	Conference Proceeding
utslib.for	0801 Artificial Intelligence and Image Processing	en_US
dc.location.activity	Melbourne, Australia	en_US
pubs.embargo.period	Not known	en_US
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	closed_access
pubs.publication-status	Published	en_US

Abstract:

Multiple Instance Learning (MIL) has been widely applied in practice, such as drug activity prediction, content-based image retrieval. In MIL, a sample, comprised of a set of instances, is called a bag. Labels are assigned to bags instead of instances. The uncertainty of labels on instances makes MIL different from conventional supervised single instance learning (SIL) tasks. Therefore, it is critical to learn an effective mapping to convert an MIL task to an SIL task. In this paper, we present OptMILES by learning the optimal transformation on the bag-to-instance similarity measure, exploring the optimal distance metric between instances, by an alternating minimization training procedure. We thoroughly evaluate the proposed method on both a synthetic dataset and real world datasets by comparing with representative MIL algorithms. The experimental results suggest the effectiveness of OptMILES. © 2013 IEEE.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/28032