CCL: Collaborative Curriculum Learning for Sparse-Reward Multi-Agent Reinforcement Learning via Co-evolutionary Task Evolution

CCL: Collaborative Curriculum Learning for Sparse-Reward Multi-Agent Reinforcement Learning via Co-evolutionary Task Evolution | 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 CCL: Collaborative Curriculum Learning for Sparse-Reward Multi-Agent Reinforcement Learning via Co-evolutionary Task Evolution Yufei Lin, Chengwei Ye, Huanzhen Zhang, Kangsheng Wang, Linuo Xu, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6605272/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Sparse reward environments pose significant challenges in reinforcement learning, especially within multi-agent systems (MAS) where feedback is delayed and shared across agents, leading to suboptimal learning. We propose Collaborative Multi-dimensional Course Learning (CCL), a novel curriculum learning framework that addresses this by (1) refining intermediate tasks for individual agents, (2) using a variational evolutionary algorithm to generate informative subtasks, and (3) co-evolving agents with their environment to enhance training stability. Experiments on five cooperative tasks in the MPE and Hide-and-Seek environments show that CCL outperforms existing methods in sparse reward settings. Multi-Agent Reinforcement Learning (MARL) Sparse Reward Environments Curriculum Learning Co-evolutionary Algorithms Task Generation Evolutionary Reinforcement Learning Cooperative Problem Solving Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted 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-6605272","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":452829601,"identity":"0d07cc7d-1d1c-4d5f-9dca-2898dd9537e3","order_by":0,"name":"Yufei Lin","email":"","orcid":"https://orcid.org/0009-0005-0941-3316","institution":"Homesite Group Inc","correspondingAuthor":false,"prefix":"","firstName":"Yufei","middleName":"","lastName":"Lin","suffix":""},{"id":452829602,"identity":"713935d2-5da0-43bc-9118-3634c3f83d09","order_by":1,"name":"Chengwei Ye","email":"","orcid":"https://orcid.org/0009-0004-2593-3621","institution":"Homesite Group Inc","correspondingAuthor":false,"prefix":"","firstName":"Chengwei","middleName":"","lastName":"Ye","suffix":""},{"id":452829603,"identity":"27b4f895-63bc-4370-9318-514e2f3feed7","order_by":2,"name":"Huanzhen Zhang","email":"","orcid":"https://orcid.org/0009-0008-3051-5642","institution":"Chewy Inc","correspondingAuthor":false,"prefix":"","firstName":"Huanzhen","middleName":"","lastName":"Zhang","suffix":""},{"id":452829604,"identity":"998e7f6e-b688-4e39-b132-ba0fa4b4fa15","order_by":3,"name":"Kangsheng Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuklEQVRIiWNgGAWjYBACxgYGZoYEBhsekrWkkaAFCJiB+DAp6tsPPzZ4UHFehp+9+QDDxz21RDisJ804IeHMbR7JnmMJjDOeHSdCS0MO84HEtts8BjdyDJh5DhwjQkv/G6CWf+d47O+//0Cklhk5zAmJDQd4DCR4GIBaaojR8szYIOFYMo/EmTSDgzMOHCCsxbA/+bHkjxo7e/72ww8ffDhQR4SWBiTOAaIiSB6NT4Qto2AUjIJRMOIAAMGAO4EAfYQ5AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0009-8392-4148","institution":"University of Science and Technology Beijing","correspondingAuthor":true,"prefix":"","firstName":"Kangsheng","middleName":"","lastName":"Wang","suffix":""},{"id":452829605,"identity":"db383672-3df6-4b6a-8c31-054a6caabf66","order_by":4,"name":"Linuo Xu","email":"","orcid":"https://orcid.org/0009-0004-2901-6193","institution":"Yunnan University of Finance and Economics","correspondingAuthor":false,"prefix":"","firstName":"Linuo","middleName":"","lastName":"Xu","suffix":""},{"id":452829606,"identity":"ea125454-30b2-4690-939c-baba9e10302c","order_by":5,"name":"Shuyan Liu","email":"","orcid":"https://orcid.org/0009-0004-7641-2623","institution":"Yunnan University","correspondingAuthor":false,"prefix":"","firstName":"Shuyan","middleName":"","lastName":"Liu","suffix":""},{"id":452829607,"identity":"c2dd3e71-723e-4f64-b040-7b471056a5e8","order_by":6,"name":"Zeyu Zhang","email":"","orcid":"","institution":"The Australian National University","correspondingAuthor":false,"prefix":"","firstName":"Zeyu","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-05-06 16:55:09","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6605272/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6605272/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82180541,"identity":"c1109427-1dfa-4c79-be12-3da0ff79a8ca","added_by":"auto","created_at":"2025-05-07 11:47:07","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1225140,"visible":true,"origin":"","legend":"","description":"","filename":"CCL.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6605272/v1_covered_d71e18a8-88fe-474e-af2b-504983c222a3.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eCCL: Collaborative Curriculum Learning for Sparse-Reward Multi-Agent Reinforcement Learning via Co-evolutionary Task Evolution\u003c/strong\u003e\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University of Science and Technology Beijing","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Multi-Agent Reinforcement Learning (MARL), Sparse Reward Environments, Curriculum Learning, Co-evolutionary Algorithms, Task Generation, Evolutionary Reinforcement Learning, Cooperative Problem Solving","lastPublishedDoi":"10.21203/rs.3.rs-6605272/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6605272/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSparse reward environments pose significant challenges in reinforcement learning, especially within multi-agent systems (MAS) where feedback is delayed and shared across agents, leading to suboptimal learning. We propose Collaborative Multi-dimensional Course Learning (CCL), a novel curriculum learning framework that addresses this by (1) refining intermediate tasks for individual agents, (2) using a variational evolutionary algorithm to generate informative subtasks, and (3) co-evolving agents with their environment to enhance training stability. Experiments on five cooperative tasks in the MPE and Hide-and-Seek environments show that CCL outperforms existing methods in sparse reward settings.\u003c/p\u003e","manuscriptTitle":"CCL: Collaborative Curriculum Learning for Sparse-Reward Multi-Agent Reinforcement Learning via Co-evolutionary Task Evolution","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-07 11:39:00","doi":"10.21203/rs.3.rs-6605272/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a6392f3e-4443-42fb-9d76-a069225b2ce8","owner":[],"postedDate":"May 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-05-07T11:39:00+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-07 11:39:00","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6605272","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6605272","identity":"rs-6605272","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}