Show simple item record

Graph Homomorphism Features: Why Not Sample?

hal.structure.identifierLaboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
dc.contributor.authorBeaujean, Paul
ORCID: 0000-0002-4707-8388
hal.structure.identifierLaboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
hal.structure.identifierLehrstuhl Bioinformatik Jena
dc.contributor.authorSikora, Florian
HAL ID: 742949
ORCID: 0000-0003-2670-6258
hal.structure.identifierLaboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
dc.contributor.authorYger, Florian
HAL ID: 17768
ORCID: 0000-0002-7182-8062
dc.date.accessioned2023-01-31T10:53:04Z
dc.date.available2023-01-31T10:53:04Z
dc.date.issued2022
dc.identifier.urihttps://basepub.dauphine.psl.eu/handle/123456789/23935
dc.language.isoenen
dc.subjectGraph embeddingen
dc.subjectGraph homomorphismen
dc.subjectGraph classificationen
dc.subject.ddc003en
dc.titleGraph Homomorphism Features: Why Not Sample?en
dc.typeCommunication / Conférence
dc.description.abstractenRecent research in the domain of computed graph embeddings has shown that graph homomorphism numbers constitute expressive features that are well-suited for machine learning tasks such as graph classification. In this work-in-progress paper, we attempt to make this methodology scalable by obtaining additive approximations to graph homomorphism densities via a simple sampling algorithm. We show in experiments that these approximate homomorphism densities perform as well as homomorphism numbers on standard graph classification datasets. Moreover, we show that, unlike algorithms that compute homomorphism numbers, our sampling algorithm is highly scalable to larger graphs.en
dc.identifier.citationpages216–222en
dc.relation.ispartoftitleMachine Learning and Principles and Practice of Knowledge Discovery in Databasesen
dc.relation.ispartofpublnameSpringer International Publishingen
dc.relation.ispartofpublcityBerlin Heidelbergen
dc.relation.ispartofdate2022-02
dc.relation.ispartofpages882en
dc.relation.ispartofurl10.1007/978-3-030-93736-2en
dc.identifier.urlsitehttps://hal.archives-ouvertes.fr/hal-03583713en
dc.subject.ddclabelRecherche opérationnelleen
dc.relation.ispartofisbn978-3-030-93735-5en
dc.relation.conftitleInternational Workshops of ECML PKDD 2021en
dc.relation.confdate2021-09
dc.relation.confcityBilbaoen
dc.relation.confcountrySpainen
dc.relation.forthcomingnonen
dc.identifier.doi10.1007/978-3-030-93736-2_17en
dc.description.ssrncandidatenon
dc.description.halcandidatenonen
dc.description.readershiprechercheen
dc.description.audienceInternationalen
dc.relation.Isversionofjnlpeerreviewedouien
dc.date.updated2023-01-31T10:40:15Z
hal.export.arxivnonen
hal.export.pmcnonen
hal.hide.repecnonen
hal.hide.oainonen
hal.author.functionaut
hal.author.functionaut
hal.author.functionaut


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record