Type of Publication: | Journal article |
Publication status: | Published |
URI (citable link): | http://nbn-resolving.de/urn:nbn:de:bsz:352-2-1gmjjj99lu4t78 |
Author: | Etheredge, Robert Ian; Schartl, Manfred; Jordan, Alex |
Year of publication: | 2021 |
Published in: | Patterns ; 2 (2021), 2. - 100193. - Cell Press. - eISSN 2666-3899 |
DOI (citable link): | https://dx.doi.org/10.1016/j.patter.2020.100193 |
Summary: |
Apart from discriminative modeling, the application of deep convolutional neural networks to basic research utilizing natural imaging data faces unique hurdles. Here, we present decontextualized hierarchical representation learning (DHRL), designed specifically to overcome these limitations. DHRL enables the broader use of small datasets, which are typical in most studies. It also captures spatial relationships between features, provides novel tools for investigating latent variables, and achieves state-of-the-art disentanglement scores on small datasets. DHRL is enabled by a novel preprocessing technique inspired by generative model chaining and an improved ladder network architecture and regularization scheme. More than an analytical tool, DHRL enables novel capabilities for virtual experiments performed directly on a latent representation, which may transform the way we perform investigations of natural image features, directly integrating analytical, empirical, and theoretical approaches.
|
Subject (DDC): | 570 Biosciences, Biology |
Link to License: | Attribution-NonCommercial-NoDerivatives 4.0 International |
Bibliography of Konstanz: | Yes |
Refereed: | Unknown |
ETHEREDGE, Robert Ian, Manfred SCHARTL, Alex JORDAN, 2021. Decontextualized learning for interpretable hierarchical representations of visual patterns. In: Patterns. Cell Press. 2(2), 100193. eISSN 2666-3899. Available under: doi: 10.1016/j.patter.2020.100193
@article{Etheredge2021-02Decon-52942, title={Decontextualized learning for interpretable hierarchical representations of visual patterns}, year={2021}, doi={10.1016/j.patter.2020.100193}, number={2}, volume={2}, journal={Patterns}, author={Etheredge, Robert Ian and Schartl, Manfred and Jordan, Alex}, note={Article Number: 100193} }
<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/rdf/resource/123456789/52942"> <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/52942"/> <dc:contributor>Etheredge, Robert Ian</dc:contributor> <dcterms:title>Decontextualized learning for interpretable hierarchical representations of visual patterns</dcterms:title> <dspace:hasBitstream rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/52942/1/Etheredge_2-1gmjjj99lu4t78.pdf"/> <dc:language>eng</dc:language> <dc:contributor>Jordan, Alex</dc:contributor> <dc:creator>Etheredge, Robert Ian</dc:creator> <dc:rights>Attribution-NonCommercial-NoDerivatives 4.0 International</dc:rights> <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/rdf/resource/123456789/28"/> <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/rdf/resource/123456789/28"/> <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2021-02-19T12:58:16Z</dcterms:available> <dcterms:abstract xml:lang="eng">Apart from discriminative modeling, the application of deep convolutional neural networks to basic research utilizing natural imaging data faces unique hurdles. Here, we present decontextualized hierarchical representation learning (DHRL), designed specifically to overcome these limitations. DHRL enables the broader use of small datasets, which are typical in most studies. It also captures spatial relationships between features, provides novel tools for investigating latent variables, and achieves state-of-the-art disentanglement scores on small datasets. DHRL is enabled by a novel preprocessing technique inspired by generative model chaining and an improved ladder network architecture and regularization scheme. More than an analytical tool, DHRL enables novel capabilities for virtual experiments performed directly on a latent representation, which may transform the way we perform investigations of natural image features, directly integrating analytical, empirical, and theoretical approaches.</dcterms:abstract> <dc:creator>Schartl, Manfred</dc:creator> <dc:contributor>Schartl, Manfred</dc:contributor> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dcterms:hasPart rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/52942/1/Etheredge_2-1gmjjj99lu4t78.pdf"/> <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2021-02-19T12:58:16Z</dc:date> <foaf:homepage rdf:resource="http://localhost:8080/jspui"/> <dc:creator>Jordan, Alex</dc:creator> <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/rdf/resource/123456789/43615"/> <dcterms:rights rdf:resource="http://creativecommons.org/licenses/by-nc-nd/4.0/"/> <dcterms:issued>2021-02</dcterms:issued> <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/rdf/resource/123456789/43615"/> </rdf:Description> </rdf:RDF>
Etheredge_2-1gmjjj99lu4t78.pdf | 155 |