Publikation: Verifying Global Neural Network Specifications using Hyperproperties
Dateien
Datum
Autor:innen
Herausgeber:innen
ISSN der Zeitschrift
Electronic ISSN
ISBN
Bibliografische Daten
Verlag
Schriftenreihe
Auflagebezeichnung
URI (zitierfähiger Link)
DOI (zitierfähiger Link)
Internationale Patentnummer
Link zur Lizenz
Angaben zur Forschungsförderung
Projekt
Open Access-Veröffentlichung
Core Facility der Universität Konstanz
Titel in einer weiteren Sprache
Publikationstyp
Publikationsstatus
Erschienen in
Zusammenfassung
Current approaches to neural network verification focus on specifications that target small regions around known input data points, such as local robustness. Thus, using these approaches, we can not obtain guarantees for inputs that are not close to known inputs. Yet, it is highly likely that a neural network will encounter such truly unseen inputs during its application. We study global specifications that — when satisfied — provide guarantees for all potential inputs. We introduce a hyperproperty formalism that allows for expressing global specifications such as monotonicity, Lipschitz continuity, global robustness, and dependency fairness. Our formalism enables verifying global specifications using existing neural network verification approaches by leveraging capabilities for verifying general computational graphs. Thereby, we extend the scope of guarantees that can be provided using existing methods. Recent success in verifying specific global specifications shows that attaining strong guarantees for all potential data points is feasible.
Zusammenfassung in einer weiteren Sprache
Fachgebiet (DDC)
Schlagwörter
Konferenz
Rezension
Zitieren
ISO 690
BOETIUS, David, Stefan LEUE, 2023. Verifying Global Neural Network Specifications using Hyperproperties. 6th Workshop on Formal Methods for ML-Enabled Autonomous Systems : FoMLAS2023. Paris, France, 17. Juli 2023 - 18. Juli 2023. In: NARODYTSKA, Nina, Hrsg., Guy AMIR, Hrsg., Guy KATZ, Hrsg. und andere. Proceedings of the 6th Workshop on Formal Methods for ML-Enabled Autonomous Systems : FoMLAS2023. EasyChair, 2023. Kalpa Publications in Computing. 16. ISSN 2515-1762. Verfügbar unter: doi: 10.29007/pvtnBibTex
@inproceedings{Boetius2023Verif-70511, year={2023}, doi={10.29007/pvtn}, title={Verifying Global Neural Network Specifications using Hyperproperties}, number={16}, issn={2515-1762}, publisher={EasyChair}, series={Kalpa Publications in Computing}, booktitle={Proceedings of the 6th Workshop on Formal Methods for ML-Enabled Autonomous Systems : FoMLAS2023}, editor={Narodytska, Nina and Amir, Guy and Katz, Guy}, author={Boetius, David and Leue, Stefan} }
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/123456789/70511"> <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2024-07-31T10:41:12Z</dc:date> <foaf:homepage rdf:resource="http://localhost:8080/"/> <dc:creator>Boetius, David</dc:creator> <dcterms:abstract>Current approaches to neural network verification focus on specifications that target small regions around known input data points, such as local robustness. Thus, using these approaches, we can not obtain guarantees for inputs that are not close to known inputs. Yet, it is highly likely that a neural network will encounter such truly unseen inputs during its application. We study global specifications that — when satisfied — provide guarantees for all potential inputs. We introduce a hyperproperty formalism that allows for expressing global specifications such as monotonicity, Lipschitz continuity, global robustness, and dependency fairness. Our formalism enables verifying global specifications using existing neural network verification approaches by leveraging capabilities for verifying general computational graphs. Thereby, we extend the scope of guarantees that can be provided using existing methods. Recent success in verifying specific global specifications shows that attaining strong guarantees for all potential data points is feasible.</dcterms:abstract> <dcterms:issued>2023</dcterms:issued> <dcterms:title>Verifying Global Neural Network Specifications using Hyperproperties</dcterms:title> <dspace:hasBitstream rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/70511/1/Boetius_2-vqdbjrm55ool7.pdf"/> <dcterms:hasPart rdf:resource="https://kops.uni-konstanz.de/bitstream/123456789/70511/1/Boetius_2-vqdbjrm55ool7.pdf"/> <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/70511"/> <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/> <dc:language>eng</dc:language> <dc:contributor>Boetius, David</dc:contributor> <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dc:creator>Leue, Stefan</dc:creator> <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2024-07-31T10:41:12Z</dcterms:available> <dc:contributor>Leue, Stefan</dc:contributor> </rdf:Description> </rdf:RDF>