Distributed Hybrid Quantum-Classical Performance Prediction for Hyperparameter Optimization

Distributed Hybrid Quantum-Classical Performance Prediction for Hyperparameter Optimization | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Distributed Hybrid Quantum-Classical Performance Prediction for Hyperparameter Optimization Eric Wulff, Juan Pablo Garcia Amboage, Marcel Aach, Thorsteinn Eli Gislason, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4270639/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 14 Sep, 2024 Read the published version in Quantum Machine Intelligence → Version 1 posted 9 You are reading this latest preprint version Abstract Hyperparameter Optimization of neural networks is a computationally expensive procedure, which requires a large number of different model configurations to be trained. To reduce such costs, this work presents a distributed, hybrid workflow, that runs the training of the neural networks on multiple Graphics Processing Units (GPUs) on a classical supercomputer, while predicting the configurations' performance with Quantum Support Vector Regression on a Quantum Annealer (QA). The workflow is shown to run on up to 50 GPUs and a QA at the same time, completely automating the communication between the classical and the quantum system. The approach is evaluated extensively on several benchmarking datasets from the Computer Vision, High Energy Physics, and Natural Language Processing domains. Results show that resource savings of up to approximately 9% can be achieved while obtaining similar, and in some cases even better, accuracy, highlighting the potential of hybrid quantum-classical machine learning algorithms. The workflow code is made available open-source to foster adoption in the community. Hyperparameter Optimization Quantum Annealing Hyperband Distributed Computing Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 14 Sep, 2024 Read the published version in Quantum Machine Intelligence → Version 1 posted Editorial decision: Revision requested 28 Jun, 2024 Reviews received at journal 13 Jun, 2024 Reviews received at journal 04 Jun, 2024 Reviewers agreed at journal 30 May, 2024 Reviewers agreed at journal 15 May, 2024 Reviewers invited by journal 14 May, 2024 Editor assigned by journal 23 Apr, 2024 Submission checks completed at journal 18 Apr, 2024 First submitted to journal 15 Apr, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4270639","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":292949224,"identity":"e5a4e014-1f64-4c2b-9968-b1cd84025bf6","order_by":0,"name":"Eric Wulff","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAw0lEQVRIiWNgGAWjYFACxgYQKcPefPgARECCSC08PMfSEojVAgFALTkGxGnhlz7c+IGx7TAPDxvPxw8f99xh4J/dgF+LZF9iswREC+9myRnPnjFI3DmAX4vBGcY2BsZth3ns5Xu3MfMcOMxgIJGAX4s9TAvQYc+Y/xCjxYAHoYWNmYEYLRJnGJslEv+lA7WwGUv2HDjMI3GDgBb+HvaHHz6csZYD2vHww48Dh+X4ZxDQAgbIaniIUD8KRsEoGAWjgBAAAJf5PFQArRTrAAAAAElFTkSuQmCC","orcid":"","institution":"European Organization for Nuclear Research","correspondingAuthor":true,"prefix":"","firstName":"Eric","middleName":"","lastName":"Wulff","suffix":""},{"id":292949225,"identity":"e5759285-b429-4da8-9497-d5caa034f2bd","order_by":1,"name":"Juan Pablo Garcia Amboage","email":"","orcid":"","institution":"European Organization for Nuclear Research","correspondingAuthor":false,"prefix":"","firstName":"Juan","middleName":"Pablo Garcia","lastName":"Amboage","suffix":""},{"id":292949226,"identity":"f6a14f22-8129-48ed-8793-6785a412c097","order_by":2,"name":"Marcel Aach","email":"","orcid":"","institution":"Forschungszentrum Jülich","correspondingAuthor":false,"prefix":"","firstName":"Marcel","middleName":"","lastName":"Aach","suffix":""},{"id":292949228,"identity":"138f005e-4993-4763-8f31-8e57e412380b","order_by":3,"name":"Thorsteinn Eli Gislason","email":"","orcid":"","institution":"University of Iceland","correspondingAuthor":false,"prefix":"","firstName":"Thorsteinn","middleName":"Eli","lastName":"Gislason","suffix":""},{"id":292949230,"identity":"707b7180-5ad2-4bfc-919f-95904300d883","order_by":4,"name":"Thorsteinn Kristinn Ingolfsson","email":"","orcid":"","institution":"University of Iceland","correspondingAuthor":false,"prefix":"","firstName":"Thorsteinn","middleName":"Kristinn","lastName":"Ingolfsson","suffix":""},{"id":292949231,"identity":"6f96e3b4-7813-4c5c-bedd-85ccbbcf53d4","order_by":5,"name":"Tomas Kristinn Ingolfsson","email":"","orcid":"","institution":"University of Iceland","correspondingAuthor":false,"prefix":"","firstName":"Tomas","middleName":"Kristinn","lastName":"Ingolfsson","suffix":""},{"id":292949232,"identity":"9dd7fbb4-e4f7-4132-a911-849d9b2b8c7a","order_by":6,"name":"Edoardo Pasetto","email":"","orcid":"","institution":"Forschungszentrum Jülich","correspondingAuthor":false,"prefix":"","firstName":"Edoardo","middleName":"","lastName":"Pasetto","suffix":""},{"id":292949233,"identity":"f68f1a60-d07c-4a08-a9fc-402b49d7111a","order_by":7,"name":"Amer Delilbasic","email":"","orcid":"","institution":"Forschungszentrum Jülich","correspondingAuthor":false,"prefix":"","firstName":"Amer","middleName":"","lastName":"Delilbasic","suffix":""},{"id":292949234,"identity":"098dadbb-71a0-448c-8562-907ae626818a","order_by":8,"name":"Morris Riedel","email":"","orcid":"","institution":"University of Iceland","correspondingAuthor":false,"prefix":"","firstName":"Morris","middleName":"","lastName":"Riedel","suffix":""},{"id":292949235,"identity":"323330aa-e315-4dbd-b900-dbde8c696b71","order_by":9,"name":"Rakesh Sarma","email":"","orcid":"","institution":"Forschungszentrum Jülich","correspondingAuthor":false,"prefix":"","firstName":"Rakesh","middleName":"","lastName":"Sarma","suffix":""},{"id":292949236,"identity":"5782adda-dc19-40d0-8a74-db946a19c21b","order_by":10,"name":"Maria Girone","email":"","orcid":"","institution":"European Organization for Nuclear Research","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"","lastName":"Girone","suffix":""},{"id":292949237,"identity":"f523388f-ddc0-48e3-93fb-030310af7d62","order_by":11,"name":"Andreas Lintermann","email":"","orcid":"","institution":"Forschungszentrum Jülich","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Lintermann","suffix":""}],"badges":[],"createdAt":"2024-04-15 15:20:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4270639/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4270639/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s42484-024-00198-5","type":"published","date":"2024-09-14T15:57:28+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":64619148,"identity":"3edd74b6-b6ac-4366-b5f1-cb3aa8aa28d0","added_by":"auto","created_at":"2024-09-16 16:11:56","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1509340,"visible":true,"origin":"","legend":"","description":"","filename":"QTML23Paper.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4270639/v1_covered_8fe5d952-4b48-4349-ae19-cf04f5872997.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Distributed Hybrid Quantum-Classical Performance Prediction for Hyperparameter Optimization","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"quantum-machine-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"qumi","sideBox":"Learn more about [Quantum Machine Intelligence](http://link.springer.com/journal/42484)","snPcode":"42484","submissionUrl":"https://submission.nature.com/new-submission/42484/3","title":"Quantum Machine Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Hyperparameter Optimization, Quantum Annealing, Hyperband, Distributed Computing","lastPublishedDoi":"10.21203/rs.3.rs-4270639/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4270639/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Hyperparameter Optimization of neural networks is a computationally expensive procedure, which requires a large number of different model configurations to be trained. To reduce such costs, this work presents a distributed, hybrid workflow, that runs the training of the neural networks on multiple Graphics Processing Units (GPUs) on a classical supercomputer, while predicting the configurations' performance with Quantum Support Vector Regression on a Quantum Annealer (QA). The workflow is shown to run on up to 50 GPUs and a QA at the same time, completely automating the communication between the classical and the quantum system. The approach is evaluated extensively on several benchmarking datasets from the Computer Vision, High Energy Physics, and Natural Language Processing domains. Results show that resource savings of up to approximately 9% can be achieved while obtaining similar, and in some cases even better, accuracy, highlighting the potential of hybrid quantum-classical machine learning algorithms. The workflow code is made available open-source to foster adoption in the community.","manuscriptTitle":"Distributed Hybrid Quantum-Classical Performance Prediction for Hyperparameter Optimization","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-23 02:36:47","doi":"10.21203/rs.3.rs-4270639/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-06-28T15:36:42+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-06-13T15:22:46+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-06-04T13:26:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"105487947697920860057619171006602455345","date":"2024-05-30T07:42:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"39447239413865415725051670112224195934","date":"2024-05-15T07:29:22+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-05-14T16:07:14+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-04-23T08:17:16+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-04-18T22:26:38+00:00","index":"","fulltext":""},{"type":"submitted","content":"Quantum Machine Intelligence","date":"2024-04-15T15:18:47+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"quantum-machine-intelligence","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"qumi","sideBox":"Learn more about [Quantum Machine Intelligence](http://link.springer.com/journal/42484)","snPcode":"42484","submissionUrl":"https://submission.nature.com/new-submission/42484/3","title":"Quantum Machine Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"a2d8706d-d78e-40ea-a4d9-ebd49301e757","owner":[],"postedDate":"April 23rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-09-16T16:02:02+00:00","versionOfRecord":{"articleIdentity":"rs-4270639","link":"https://doi.org/10.1007/s42484-024-00198-5","journal":{"identity":"quantum-machine-intelligence","isVorOnly":false,"title":"Quantum Machine Intelligence"},"publishedOn":"2024-09-14 15:57:28","publishedOnDateReadable":"September 14th, 2024"},"versionCreatedAt":"2024-04-23 02:36:47","video":"","vorDoi":"10.1007/s42484-024-00198-5","vorDoiUrl":"https://doi.org/10.1007/s42484-024-00198-5","workflowStages":[]},"version":"v1","identity":"rs-4270639","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4270639","identity":"rs-4270639","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}