Rethinking and Recomputing the Value of Machine Learning Models

Rethinking and Recomputing the Value of Machine Learning Models | 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 Rethinking and Recomputing the Value of Machine Learning Models Burcu Sayin, Jie Yang, Xinyue Chen, Andrea Passerini, Fabio Casati This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4833578/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 May, 2025 Read the published version in Artificial Intelligence Review → Version 1 posted 13 You are reading this latest preprint version Abstract In this paper, we contend that the conventional approach to training and evaluating machine learning models frequently overlooks their application within the real-world organizational or societal contexts they are meant to serve. By shifting to this perspective, we redefine how we assess and choose machine learning models. Our focus is particularly on integrating these models into practical workflows that involve both machines and human experts, with human intervention occurring when machines lack sufficient confidence in their predictions. We demonstrate that traditional metrics such as accuracy and f-score fall short in capturing the true value of machine learning models in such hybrid settings. To address this issue, we introduce a simple but theoretically sound strategy to adapt existing machine learning models so as to maximize value. An extensive experimental evaluation highlights the importance of the value-based perspective in evaluating models, and the impact of calibration and out-of-distribution settings on model value. machine learning hybrid intelligence selective classification cost-sensitive learning Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 08 May, 2025 Read the published version in Artificial Intelligence Review → Version 1 posted Editorial decision: Revision requested 15 Nov, 2024 Reviews received at journal 10 Nov, 2024 Reviews received at journal 25 Oct, 2024 Reviews received at journal 17 Oct, 2024 Reviewers agreed at journal 05 Oct, 2024 Reviewers agreed at journal 05 Oct, 2024 Reviewers agreed at journal 04 Oct, 2024 Reviewers agreed at journal 03 Oct, 2024 Reviewers agreed at journal 28 Aug, 2024 Reviewers invited by journal 23 Aug, 2024 Editor assigned by journal 09 Aug, 2024 Submission checks completed at journal 31 Jul, 2024 First submitted to journal 31 Jul, 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-4833578","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":344877120,"identity":"8e9df928-2144-429a-87d3-fcd501dae704","order_by":0,"name":"Burcu Sayin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1klEQVRIiWNgGAWjYBACAziLmYHxAQODBJDF2EC0FmYgRwKoh7ERrx4DJDYbyArC1phL5D78zMNQJ29wnPdYNe8OizpzBub2B/i0WM5IN5bmYThsuOEwX9pt3jMSEpYNhBx2I42NmYfhAOOGwzxmt3nbJCQMDhCnpc4epKWYFC3MiSAtzERpsex5xiw5x+Bw8szDPMaSc9skJDccZmycgU+LOXsa44c3FXW2fefPGH5421bHb3C8/cEHfFqgzmNgUDgA4zATVg8B8g3EqhwFo2AUjIIRBwAEi0EZ1c1OPQAAAABJRU5ErkJggg==","orcid":"","institution":"University of Trento","correspondingAuthor":true,"prefix":"","firstName":"Burcu","middleName":"","lastName":"Sayin","suffix":""},{"id":344877121,"identity":"358011c1-cae6-43ac-b94f-60fa08fd4d12","order_by":1,"name":"Jie Yang","email":"","orcid":"","institution":"Delft University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Yang","suffix":""},{"id":344877122,"identity":"b53e5002-d4c4-4944-b8fb-771f76b93852","order_by":2,"name":"Xinyue Chen","email":"","orcid":"","institution":"Delft University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Xinyue","middleName":"","lastName":"Chen","suffix":""},{"id":344877123,"identity":"77d01c47-ba04-4fdf-b790-6ce0242fe230","order_by":3,"name":"Andrea Passerini","email":"","orcid":"","institution":"University of Trento","correspondingAuthor":false,"prefix":"","firstName":"Andrea","middleName":"","lastName":"Passerini","suffix":""},{"id":344877124,"identity":"9ed13b2b-f912-4141-9855-e01c319815e6","order_by":4,"name":"Fabio Casati","email":"","orcid":"","institution":"ServiceNow (United States)","correspondingAuthor":false,"prefix":"","firstName":"Fabio","middleName":"","lastName":"Casati","suffix":""}],"badges":[],"createdAt":"2024-07-31 08:17:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4833578/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4833578/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10462-025-11242-6","type":"published","date":"2025-05-08T15:57:10+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":82537488,"identity":"a90a6580-93b4-4a25-883d-b5d7ebaea4b2","added_by":"auto","created_at":"2025-05-12 16:07:22","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":730002,"visible":true,"origin":"","legend":"","description":"","filename":"RethinkingandRecomputingtheValueofMachineLearningModels.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4833578/v1_covered_b8c83c45-8d07-49cf-94af-a5c10638c0ef.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Rethinking and Recomputing the Value of Machine Learning Models","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":"artificial-intelligence-review","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"aire","sideBox":"Learn more about [Artificial Intelligence Review](http://link.springer.com/journal/10462)","snPcode":"10462","submissionUrl":"https://submission.nature.com/new-submission/10462/3","title":"Artificial Intelligence Review","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"machine learning, hybrid intelligence, selective classification, cost-sensitive learning","lastPublishedDoi":"10.21203/rs.3.rs-4833578/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4833578/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"In this paper, we contend that the conventional approach to training and evaluating machine learning models frequently overlooks their application within the real-world organizational or societal contexts they are meant to serve. By shifting to this perspective, we redefine how we assess and choose machine learning models. Our focus is particularly on integrating these models into practical workflows that involve both machines and human experts, with human intervention occurring when machines lack sufficient confidence in their predictions. We demonstrate that traditional metrics such as accuracy and f-score fall short in capturing the true value of machine learning models in such hybrid settings. To address this issue, we introduce a simple but theoretically sound strategy to adapt existing machine learning models so as to maximize value. An extensive experimental evaluation highlights the importance of the value-based perspective in evaluating models, and the impact of calibration and out-of-distribution settings on model value.","manuscriptTitle":"Rethinking and Recomputing the Value of Machine Learning Models","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-27 16:30:50","doi":"10.21203/rs.3.rs-4833578/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-11-15T14:28:14+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-11-10T10:36:00+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-25T23:18:01+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-17T19:50:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"40614091665225356768235113360579049354","date":"2024-10-05T07:39:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"177505631208572906023649542570262998532","date":"2024-10-05T06:36:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"217070841199633964413340837206352206976","date":"2024-10-04T21:40:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"279682836820400373618912653774448432368","date":"2024-10-04T01:45:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"305798766689232721840315349355501925467","date":"2024-08-28T12:18:29+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-08-23T07:06:15+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-08-09T07:27:28+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-07-31T14:00:17+00:00","index":"","fulltext":""},{"type":"submitted","content":"Artificial Intelligence Review","date":"2024-07-31T08:15:51+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"artificial-intelligence-review","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"aire","sideBox":"Learn more about [Artificial Intelligence Review](http://link.springer.com/journal/10462)","snPcode":"10462","submissionUrl":"https://submission.nature.com/new-submission/10462/3","title":"Artificial Intelligence Review","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"f362426a-f4a8-49b0-aa4e-daf375f44e6c","owner":[],"postedDate":"August 27th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-05-12T16:00:00+00:00","versionOfRecord":{"articleIdentity":"rs-4833578","link":"https://doi.org/10.1007/s10462-025-11242-6","journal":{"identity":"artificial-intelligence-review","isVorOnly":false,"title":"Artificial Intelligence Review"},"publishedOn":"2025-05-08 15:57:10","publishedOnDateReadable":"May 8th, 2025"},"versionCreatedAt":"2024-08-27 16:30:50","video":"","vorDoi":"10.1007/s10462-025-11242-6","vorDoiUrl":"https://doi.org/10.1007/s10462-025-11242-6","workflowStages":[]},"version":"v1","identity":"rs-4833578","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4833578","identity":"rs-4833578","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}