000866292 001__ 866292 000866292 005__ 20220930130222.0 000866292 0247_ $$2doi$$a10.1016/j.cpc.2019.107006 000866292 0247_ $$2ISSN$$a0010-4655 000866292 0247_ $$2ISSN$$a1386-9485 000866292 0247_ $$2ISSN$$a1879-2944 000866292 0247_ $$2Handle$$a2128/23791 000866292 0247_ $$2WOS$$aWOS:000509613900006 000866292 0247_ $$2altmetric$$aaltmetric:62281769 000866292 037__ $$aFZJ-2019-05451 000866292 082__ $$a530 000866292 1001_ $$0P:(DE-Juel1)167542$$aWillsch, D.$$b0$$eCorresponding author 000866292 245__ $$aSupport vector machines on the D-Wave quantum annealer 000866292 260__ $$aAmsterdam$$bNorth Holland Publ. Co.$$c2020 000866292 3367_ $$2DRIVER$$aarticle 000866292 3367_ $$2DataCite$$aOutput Types/Journal article 000866292 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1610983008_11031 000866292 3367_ $$2BibTeX$$aARTICLE 000866292 3367_ $$2ORCID$$aJOURNAL_ARTICLE 000866292 3367_ $$00$$2EndNote$$aJournal Article 000866292 520__ $$aKernel-based support vector machines (SVMs) are supervised machine learning algorithms for classification and regression problems. We introduce a method to train SVMs on a D-Wave 2000Q quantum annealer and study its performance in comparison to SVMs trained on conventional computers. The method is applied to both synthetic data and real data obtained from biology experiments. We find that the quantum annealer produces an ensemble of different solutions that often generalizes better to unseen data than the single global minimum of an SVM trained on a conventional computer, especially in cases where only limited training data is available. For cases with more training data than currently fits on the quantum annealer, we show that a combination of classifiers for subsets of the data almost always produces stronger joint classifiers than the conventional SVM for the same parameters. 000866292 536__ $$0G:(DE-HGF)POF3-511$$a511 - Computational Science and Mathematical Methods (POF3-511)$$cPOF3-511$$fPOF III$$x0 000866292 536__ $$0G:(DE-Juel1)PHD-NO-GRANT-20170405$$aPhD no Grant - Doktorand ohne besondere Förderung (PHD-NO-GRANT-20170405)$$cPHD-NO-GRANT-20170405$$x1 000866292 588__ $$aDataset connected to CrossRef 000866292 7001_ $$0P:(DE-Juel1)167543$$aWillsch, M.$$b1 000866292 7001_ $$0P:(DE-HGF)0$$aDe Raedt, Hans$$b2 000866292 7001_ $$0P:(DE-Juel1)138295$$aMichielsen, K.$$b3 000866292 773__ $$0PERI:(DE-600)1466511-6$$a10.1016/j.cpc.2019.107006$$gp. 107006 -$$p107006 -$$tComputer physics communications$$v248$$x0010-4655$$y2020 000866292 8564_ $$uhttps://juser.fz-juelich.de/record/866292/files/Rechnung-Elsevier-Willsch-CommCompPhys-2019-11-11.pdf 000866292 8564_ $$uhttps://juser.fz-juelich.de/record/866292/files/1-s2.0-S001046551930342X-main.pdf$$yOpenAccess 000866292 8564_ $$uhttps://juser.fz-juelich.de/record/866292/files/Rechnung-Elsevier-Willsch-CommCompPhys-2019-11-11.pdf?subformat=pdfa$$xpdfa 000866292 8564_ $$uhttps://juser.fz-juelich.de/record/866292/files/1-s2.0-S001046551930342X-main.pdf?subformat=pdfa$$xpdfa$$yOpenAccess 000866292 8767_ $$8OAD0000018355$$92019-11-06$$d2019-11-11$$eHybrid-OA$$jZahlung erfolgt 000866292 909CO $$ooai:juser.fz-juelich.de:866292$$popenCost$$pVDB$$pdriver$$pOpenAPC$$popen_access$$popenaire$$pdnbdelivery 000866292 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)167542$$aForschungszentrum Jülich$$b0$$kFZJ 000866292 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)167543$$aForschungszentrum Jülich$$b1$$kFZJ 000866292 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)138295$$aForschungszentrum Jülich$$b3$$kFZJ 000866292 9131_ $$0G:(DE-HGF)POF3-511$$1G:(DE-HGF)POF3-510$$2G:(DE-HGF)POF3-500$$3G:(DE-HGF)POF3$$4G:(DE-HGF)POF$$aDE-HGF$$bKey Technologies$$lSupercomputing & Big Data$$vComputational Science and Mathematical Methods$$x0 000866292 9141_ $$y2019 000866292 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS 000866292 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search 000866292 915__ $$0LIC:(DE-HGF)CCBYNCND4$$2HGFVOC$$aCreative Commons Attribution-NonCommercial-NoDerivs CC BY-NC-ND 4.0 000866292 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bCOMPUT PHYS COMMUN : 2017 000866292 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection 000866292 915__ $$0StatID:(DE-HGF)0110$$2StatID$$aWoS$$bScience Citation Index 000866292 915__ $$0StatID:(DE-HGF)0111$$2StatID$$aWoS$$bScience Citation Index Expanded 000866292 915__ $$0StatID:(DE-HGF)9900$$2StatID$$aIF < 5 000866292 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess 000866292 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC 000866292 915__ $$0StatID:(DE-HGF)1150$$2StatID$$aDBCoverage$$bCurrent Contents - Physical, Chemical and Earth Sciences 000866292 915__ $$0StatID:(DE-HGF)0310$$2StatID$$aDBCoverage$$bNCBI Molecular Biology Database 000866292 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline 000866292 915__ $$0StatID:(DE-HGF)0420$$2StatID$$aNationallizenz 000866292 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List 000866292 920__ $$lyes 000866292 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0 000866292 980__ $$ajournal 000866292 980__ $$aVDB 000866292 980__ $$aI:(DE-Juel1)JSC-20090406 000866292 980__ $$aAPC 000866292 980__ $$aUNRESTRICTED 000866292 9801_ $$aAPC 000866292 9801_ $$aFullTexts