Neural diversity is key to collective artificial learning

Neural diversity is key to collective artificial learning | 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 Physical Sciences - Article Neural diversity is key to collective artificial learning Matteo Bettini, Ryan Kortvelesy, Amanda Prorok This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5589114/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Many of the world’s most pressing issues, such as climate change and global peace, require complex collective problem-solving skills. Recent studies indicate that diversity in individuals’ behaviors is key to developing such skills and increasing collective performance. Yet behavioral diversity in collective artificial learning is understudied, with today’s machine learning paradigms commonly favoring homogeneous agent strategies over heterogeneous ones, mainly due to computational considerations. In this work, we employ novel diversity measurement and control paradigms to study the impact of behavioral heterogeneity in several facets of collective artificial learning. Through experiments in team play and other cooperative tasks, we show the emergence of unbiased behavioral roles that improve team outcomes; how neural diversity synergizes with morphological diversity; how diverse agents are more effective at finding cooperative solutions in sparse reward settings; and how behaviorally heterogeneous teams learn and retain latent skills to overcome repeated disruptions. Overall, our results indicate that, by controlling diversity, we can obtain non-trivial benefits over homogeneous training paradigms, demonstrating that diversity is a fundamental component of collective artificial learning, an insight thus far overlooked. Physical sciences/Mathematics and computing/Computer science Physical sciences/Mathematics and computing/Computational science Physical sciences/Mathematics and computing/Software Diversity Multi-Agent Learning Soccer Full Text Additional Declarations There is NO Competing Interest. Supplementary Files Trailerdiversityiskey.mp4 Soccer trailer video DIversityiskeytocollectiveartificiallearning.mp4 Extended video abstract Cite Share Download PDF Status: Under Review Version 1 posted 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-5589114","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Physical Sciences - Article","associatedPublications":[],"authors":[{"id":390076945,"identity":"7c2075dd-6b9c-4f18-a070-df428b84667f","order_by":0,"name":"Matteo Bettini","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2ElEQVRIie3RMQrCMBSA4QeB5xJ1LXToCYQEwUXxLJFCXDKL4GBByOqqeJlCQRfRtZAOFcG56OhiOunU6OaQfwgh5CMhAfD5/rGAZGU1H0HnvYQugpJvjvJzn5NQFrZ19gPp7RLGt/ocIZDLPYdxBIEUjWRQpKKsToZrwH6oIOZJINNmkouUb2ZGIFCwhAgIpomDTJKQ4qkm5Klg+Q2JIaQ6rQnaUzJLXBcr9mgfOeaa4GCo2IFrehPNxKwf9ivHUbe1uho1X9iJZI3kHakH5v5In8/n833RC62iPtZJCyqcAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-8679-0151","institution":"University of Cambridge","correspondingAuthor":true,"prefix":"","firstName":"Matteo","middleName":"","lastName":"Bettini","suffix":""},{"id":390076946,"identity":"1527e6b3-d5a9-435e-88ab-08d077c91556","order_by":1,"name":"Ryan Kortvelesy","email":"","orcid":"","institution":"University of Cambridge","correspondingAuthor":false,"prefix":"","firstName":"Ryan","middleName":"","lastName":"Kortvelesy","suffix":""},{"id":390076947,"identity":"49048f3e-5615-4a29-939c-c1a0c65b9278","order_by":2,"name":"Amanda Prorok","email":"","orcid":"","institution":"University of Cambridge","correspondingAuthor":false,"prefix":"","firstName":"Amanda","middleName":"","lastName":"Prorok","suffix":""}],"badges":[],"createdAt":"2024-12-05 19:00:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5589114/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5589114/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":74976568,"identity":"3eafb429-0035-4e6b-9a67-3fc49e406248","added_by":"auto","created_at":"2025-01-29 03:14:40","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3346956,"visible":true,"origin":"","legend":"Article File","description":"","filename":"Diversityiskeytocollectiveartificiallearning.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5589114/v1_covered_a4aa5b52-84fe-4c6a-ba9c-5c7ff225426d.pdf"},{"id":74975911,"identity":"2cfa9c05-e4e1-4684-87c0-23e3d07d8198","added_by":"auto","created_at":"2025-01-29 03:06:36","extension":"mp4","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":107293045,"visible":true,"origin":"","legend":"Soccer trailer video","description":"","filename":"Trailerdiversityiskey.mp4","url":"https://assets-eu.researchsquare.com/files/rs-5589114/v1/ccbfa32bb60ebfc5abc662fd.mp4"},{"id":74975912,"identity":"2bd9565e-8595-4f4c-96f7-c101397e4e88","added_by":"auto","created_at":"2025-01-29 03:06:38","extension":"mp4","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":237067831,"visible":true,"origin":"","legend":"Extended video abstract","description":"","filename":"DIversityiskeytocollectiveartificiallearning.mp4","url":"https://assets-eu.researchsquare.com/files/rs-5589114/v1/7f7d08273ac335398f420416.mp4"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Neural diversity is key to collective artificial learning","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Diversity, Multi-Agent, Learning, Soccer","lastPublishedDoi":"10.21203/rs.3.rs-5589114/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5589114/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Many of the world’s most pressing issues, such as climate change and global peace, require complex collective problem-solving skills. Recent studies indicate that diversity in individuals’ behaviors is key to developing such skills and increasing collective performance. Yet behavioral diversity in collective artificial learning is understudied, with today’s machine learning paradigms commonly favoring homogeneous agent strategies over heterogeneous ones, mainly due to computational considerations. In this work, we employ novel diversity measurement and control paradigms to study the impact of behavioral heterogeneity in several facets of collective artificial learning. Through experiments in team play and other cooperative tasks, we show the emergence of unbiased behavioral roles that improve team outcomes; how neural diversity synergizes with morphological diversity; how diverse agents are more effective at finding cooperative solutions in sparse reward settings; and how behaviorally heterogeneous teams learn and retain latent skills to overcome repeated disruptions. Overall, our results indicate that, by controlling diversity, we can obtain non-trivial benefits over homogeneous training paradigms, demonstrating that diversity is a fundamental component of collective artificial learning, an insight thus far overlooked.","manuscriptTitle":"Neural diversity is key to collective artificial learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-29 03:06:28","doi":"10.21203/rs.3.rs-5589114/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-machine-intelligence","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"natmachintell","sideBox":"Learn more about [Nature Machine Intelligence](http://www.nature.com/natmachintell/)","snPcode":"","submissionUrl":"","title":"Nature Machine Intelligence","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"49b03602-476b-41cb-bf61-88b50b092919","owner":[],"postedDate":"January 29th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":41581527,"name":"Physical sciences/Mathematics and computing/Computer science"},{"id":41581528,"name":"Physical sciences/Mathematics and computing/Computational science"},{"id":41581529,"name":"Physical sciences/Mathematics and computing/Software"}],"tags":[],"updatedAt":"2025-01-29T03:06:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-01-29 03:06:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5589114","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5589114","identity":"rs-5589114","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}