Differentiable Sorting Networks for Scalable Sorting and Ranking Supervision

No Thumbnail Available
Files
There are no files associated with this item.
Date
2021
Authors
Kuehne, Hilde
Editors
Contact
Journal ISSN
Electronic ISSN
ISBN
Bibliographical data
Publisher
Series
URI (citable link)
DOI (citable link)
ArXiv-ID
International patent number
Link to the license
oops
EU project number
Project
Open Access publication
Collections
Informatik und Informationswissenschaft
Restricted until
Title in another language
Research Projects
Organizational Units
Journal Issue
Publication type
Contribution to a conference collection
Publication status
Published
Published in
Proceedings of the 38th International Conference on Machine Learning / Meila, Marina; Zhang, Tong (ed.). - Maastricht : ML Research Press, 2021. - (Proceedings of Machine Learning Research ; 139). - pp. 8546-8555. - eISSN 2640-3498
Abstract
Sorting and ranking supervision is a method for training neural networks end-to-end based on ordering constraints. That is, the ground truth order of sets of samples is known, while their absolute values remain unsupervised. For that, we propose differentiable sorting networks by relaxing their pairwise conditional swap operations. To address the problems of vanishing gradients and extensive blurring that arise with larger numbers of layers, we propose mapping activations to regions with moderate gradients. We consider odd-even as well as bitonic sorting networks, which outperform existing relaxations of the sorting operation. We show that bitonic sorting networks can achieve stable training on large input sets of up to 1024 elements.
Summary in another language
Subject (DDC)
004 Computer Science
Keywords
Conference
38th International Conference on Machine Learning (virtual), Jul 18, 2021 - Jul 24, 2021
Review
undefined / . - undefined, undefined. - (undefined; undefined)
Cite This
ISO 690PETERSEN, Felix, Christian BORGELT, Hilde KUEHNE, Oliver DEUSSEN, 2021. Differentiable Sorting Networks for Scalable Sorting and Ranking Supervision. 38th International Conference on Machine Learning (virtual), Jul 18, 2021 - Jul 24, 2021. In: MEILA, Marina, ed., Tong ZHANG, ed.. Proceedings of the 38th International Conference on Machine Learning. Maastricht:ML Research Press, pp. 8546-8555. eISSN 2640-3498
BibTex
@inproceedings{Petersen2021Diffe-57905,
  year={2021},
  title={Differentiable Sorting Networks for Scalable Sorting and Ranking Supervision},
  url={http://proceedings.mlr.press/v139/petersen21a.html},
  number={139},
  publisher={ML Research Press},
  address={Maastricht},
  series={Proceedings of Machine Learning Research},
  booktitle={Proceedings of the 38th International Conference on Machine Learning},
  pages={8546--8555},
  editor={Meila, Marina and Zhang, Tong},
  author={Petersen, Felix and Borgelt, Christian and Kuehne, Hilde and Deussen, Oliver}
}
RDF
<rdf:RDF
    xmlns:dcterms="http://purl.org/dc/terms/"
    xmlns:dc="http://purl.org/dc/elements/1.1/"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:bibo="http://purl.org/ontology/bibo/"
    xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#"
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:void="http://rdfs.org/ns/void#"
    xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > 
  <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/57905">
    <dc:contributor>Deussen, Oliver</dc:contributor>
    <dc:language>eng</dc:language>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:contributor>Borgelt, Christian</dc:contributor>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2022-06-30T13:12:46Z</dcterms:available>
    <dc:contributor>Petersen, Felix</dc:contributor>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
    <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/57905"/>
    <dc:contributor>Kuehne, Hilde</dc:contributor>
    <dc:creator>Borgelt, Christian</dc:creator>
    <dc:creator>Petersen, Felix</dc:creator>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2022-06-30T13:12:46Z</dc:date>
    <dcterms:issued>2021</dcterms:issued>
    <dc:creator>Kuehne, Hilde</dc:creator>
    <dcterms:abstract xml:lang="eng">Sorting and ranking supervision is a method for training neural networks end-to-end based on ordering constraints. That is, the ground truth order of sets of samples is known, while their absolute values remain unsupervised. For that, we propose differentiable sorting networks by relaxing their pairwise conditional swap operations. To address the problems of vanishing gradients and extensive blurring that arise with larger numbers of layers, we propose mapping activations to regions with moderate gradients. We consider odd-even as well as bitonic sorting networks, which outperform existing relaxations of the sorting operation. We show that bitonic sorting networks can achieve stable training on large input sets of up to 1024 elements.</dcterms:abstract>
    <dcterms:title>Differentiable Sorting Networks for Scalable Sorting and Ranking Supervision</dcterms:title>
    <dc:creator>Deussen, Oliver</dc:creator>
  </rdf:Description>
</rdf:RDF>
Internal note
xmlui.Submission.submit.DescribeStep.inputForms.label.kops_note_fromSubmitter
Contact
URL of original publication
http://proceedings.mlr.press/v139/petersen21a.html
Test date of URL
2022-05-23
Examination date of dissertation
Method of financing
Comment on publication
Alliance license
Corresponding Authors der Uni Konstanz vorhanden
International Co-Authors
Bibliography of Konstanz
Yes
Refereed
Unknown