Unsupervised metric learning with synthetic examples
Distance Metric Learning (DML) involves learning an embedding that brings similar examples closer while moving away dissimilar ones. Existing DML approaches make use of class labels to generate constraints for metric learning. In this paper, we address the less-studied problem of learning a metric i...
Published in: | Proceedings of the AAAI Conference on Artificial Intelligence |
---|---|
Main Authors: | , , |
Other Authors: | , |
Format: | Other Non-Article Part of Journal/Newspaper |
Language: | English |
Published: |
Association for the Advancement of Artificial Intelligence (AAAI)
2020
|
Subjects: | |
Online Access: | https://research.monash.edu/en/publications/57ff31f8-f673-4448-8a1d-1a51ede2255d https://doi.org/10.1609/aaai.v34i04.5795 https://researchmgt.monash.edu/ws/files/381951922/351816113_oa.pdf http://www.scopus.com/inward/record.url?scp=85106445643&partnerID=8YFLogxK |
id |
ftmonashunicris:oai:monash.edu:publications/57ff31f8-f673-4448-8a1d-1a51ede2255d |
---|---|
record_format |
openpolar |
spelling |
ftmonashunicris:oai:monash.edu:publications/57ff31f8-f673-4448-8a1d-1a51ede2255d 2023-05-15T16:01:24+02:00 Unsupervised metric learning with synthetic examples Dutta, Ujjal K. Harandi, Mehrtash Sekhar, C. Chandra Conitzer, Vincent Sha, Fei 2020 application/pdf https://research.monash.edu/en/publications/57ff31f8-f673-4448-8a1d-1a51ede2255d https://doi.org/10.1609/aaai.v34i04.5795 https://researchmgt.monash.edu/ws/files/381951922/351816113_oa.pdf http://www.scopus.com/inward/record.url?scp=85106445643&partnerID=8YFLogxK eng eng Association for the Advancement of Artificial Intelligence (AAAI) info:eu-repo/semantics/openAccess Dutta , U K , Harandi , M & Sekhar , C C 2020 , Unsupervised metric learning with synthetic examples . in V Conitzer & F Sha (eds) , Proceedings of The Thirty-Fourth AAAI Conference on Artificial Intelligence . AAAI 2020 - 34th AAAI Conference on Artificial Intelligence , no. 4 , vol. 34 , Association for the Advancement of Artificial Intelligence (AAAI) , Palo Alto CA USA , pp. 3834-3841 , AAAI Conference on Artificial Intelligence 2020 , New York , New York , United States of America , 7/02/20 . https://doi.org/10.1609/aaai.v34i04.5795 contributionToPeriodical 2020 ftmonashunicris https://doi.org/10.1609/aaai.v34i04.5795 2023-02-05T06:40:00Z Distance Metric Learning (DML) involves learning an embedding that brings similar examples closer while moving away dissimilar ones. Existing DML approaches make use of class labels to generate constraints for metric learning. In this paper, we address the less-studied problem of learning a metric in an unsupervised manner. We do not make use of class labels, but use unlabeled data to generate adversarial, synthetic constraints for learning a metric inducing embedding. Being a measure of uncertainty, we minimize the entropy of a conditional probability to learn the metric. Our stochastic formulation scales well to large datasets, and performs competitive to existing metric learning methods. Other Non-Article Part of Journal/Newspaper DML Monash University Research Portal Proceedings of the AAAI Conference on Artificial Intelligence 34 04 3834 3841 |
institution |
Open Polar |
collection |
Monash University Research Portal |
op_collection_id |
ftmonashunicris |
language |
English |
description |
Distance Metric Learning (DML) involves learning an embedding that brings similar examples closer while moving away dissimilar ones. Existing DML approaches make use of class labels to generate constraints for metric learning. In this paper, we address the less-studied problem of learning a metric in an unsupervised manner. We do not make use of class labels, but use unlabeled data to generate adversarial, synthetic constraints for learning a metric inducing embedding. Being a measure of uncertainty, we minimize the entropy of a conditional probability to learn the metric. Our stochastic formulation scales well to large datasets, and performs competitive to existing metric learning methods. |
author2 |
Conitzer, Vincent Sha, Fei |
format |
Other Non-Article Part of Journal/Newspaper |
author |
Dutta, Ujjal K. Harandi, Mehrtash Sekhar, C. Chandra |
spellingShingle |
Dutta, Ujjal K. Harandi, Mehrtash Sekhar, C. Chandra Unsupervised metric learning with synthetic examples |
author_facet |
Dutta, Ujjal K. Harandi, Mehrtash Sekhar, C. Chandra |
author_sort |
Dutta, Ujjal K. |
title |
Unsupervised metric learning with synthetic examples |
title_short |
Unsupervised metric learning with synthetic examples |
title_full |
Unsupervised metric learning with synthetic examples |
title_fullStr |
Unsupervised metric learning with synthetic examples |
title_full_unstemmed |
Unsupervised metric learning with synthetic examples |
title_sort |
unsupervised metric learning with synthetic examples |
publisher |
Association for the Advancement of Artificial Intelligence (AAAI) |
publishDate |
2020 |
url |
https://research.monash.edu/en/publications/57ff31f8-f673-4448-8a1d-1a51ede2255d https://doi.org/10.1609/aaai.v34i04.5795 https://researchmgt.monash.edu/ws/files/381951922/351816113_oa.pdf http://www.scopus.com/inward/record.url?scp=85106445643&partnerID=8YFLogxK |
genre |
DML |
genre_facet |
DML |
op_source |
Dutta , U K , Harandi , M & Sekhar , C C 2020 , Unsupervised metric learning with synthetic examples . in V Conitzer & F Sha (eds) , Proceedings of The Thirty-Fourth AAAI Conference on Artificial Intelligence . AAAI 2020 - 34th AAAI Conference on Artificial Intelligence , no. 4 , vol. 34 , Association for the Advancement of Artificial Intelligence (AAAI) , Palo Alto CA USA , pp. 3834-3841 , AAAI Conference on Artificial Intelligence 2020 , New York , New York , United States of America , 7/02/20 . https://doi.org/10.1609/aaai.v34i04.5795 |
op_rights |
info:eu-repo/semantics/openAccess |
op_doi |
https://doi.org/10.1609/aaai.v34i04.5795 |
container_title |
Proceedings of the AAAI Conference on Artificial Intelligence |
container_volume |
34 |
container_issue |
04 |
container_start_page |
3834 |
op_container_end_page |
3841 |
_version_ |
1766397284564598784 |