Model-based Adversarial Imitation Learning with Self-adatpive Error Control

Model-based Adversarial Imitation Learning with Self-adatpive Error Control | 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 Model-based Adversarial Imitation Learning with Self-adatpive Error Control Yi Wang, Shengrong Gong, Xin Du, Shan Zhong, Yali Si This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4263827/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 Generative Adversarial Imitation Learning (GAIL) presents the ability to learn policies without prior knowledge of the underlying reward function. However, it often suffers from limited sample efficiency due to its reliance on reinforcement learning for policy learning, mandating extensive real-time interactions with the environment. To address this challenge, this paper introduces a refined framework named TM-GAIL, which combines transition function model learning with GAIL. This approach capitalizes on the utility of neural networks to construct a transition function model, facilitating the generation of virtual samples to complement real data. The training of the discriminator is augmented by the inclusion of virtual samples alongside expert demonstration data. In the context of policy learning, the incorporation of virtual samples, real samples, and the reward derived from the discriminator enriches the policy learning. Furthermore, a self-adaptive error control module has been meticulously designed for the regions characterized by high returns and to mitigate model errors. Empirical findings demonstrate that TM-GAIL significantly improves sample efficiency in comparison to imitation learning and model-free methods. It achieves performance levels that closely align with those of domain experts across both continuous and discrete tasks. Behavior Cloning Reinforcement Learning Adversarial Imitation Learning Dynamics Models Full Text Additional Declarations No competing interests reported. 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-4263827","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":291862395,"identity":"a1fcac2d-c21a-4187-96e4-9ba4c40a1029","order_by":0,"name":"Yi Wang","email":"","orcid":"","institution":"Suzhou University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Wang","suffix":""},{"id":291862396,"identity":"b7e45bc0-8d04-418c-b716-5c34599ab22d","order_by":1,"name":"Shengrong Gong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2klEQVRIiWNgGAWjYDACCQglB8QGQMxMvBZj0rUkNhCtRX527+MXH3fUps+fkbzxA0OFdWID+9kDeLUY3DluZjnzzPHcDTfSiiUYzqQnNvDkJeDXIpHGZszbdix3g0SOgQRj2+HEBgkeA/wOmwHRki4/I8f4B+M/IrQw3EhjfszbVpPAcCPHTIKxgQgtBjfS2Bhnth0w3HDmWZlFwrF04zaeHIIOY/7wsa1OXr49efONDzXWsv3sZwg4jIGBDRg1hyHMBBCXkHogYP7AwFBHhLpRMApGwSgYsQAAVxJFQuomzxoAAAAASUVORK5CYII=","orcid":"","institution":"Changshu Institute of Technology","correspondingAuthor":true,"prefix":"","firstName":"Shengrong","middleName":"","lastName":"Gong","suffix":""},{"id":291862397,"identity":"54bbdd18-4ce7-4fe5-8058-808f61d60a6d","order_by":2,"name":"Xin Du","email":"","orcid":"","institution":"Suzhou University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Xin","middleName":"","lastName":"Du","suffix":""},{"id":291862398,"identity":"ea8b1065-85f6-4076-a81b-9b84847e491c","order_by":3,"name":"Shan Zhong","email":"","orcid":"","institution":"Changshu Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Shan","middleName":"","lastName":"Zhong","suffix":""},{"id":291862399,"identity":"a614b724-224e-4966-bd6c-b8a5d6054c54","order_by":4,"name":"Yali Si","email":"","orcid":"","institution":"Changshu Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Yali","middleName":"","lastName":"Si","suffix":""}],"badges":[],"createdAt":"2024-04-14 06:44:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4263827/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4263827/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":84957384,"identity":"e1a04081-04a0-4d62-8912-692dd07d8ef7","added_by":"auto","created_at":"2025-06-19 08:24:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4360862,"visible":true,"origin":"","legend":"","description":"","filename":"Template.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4263827/v1_covered_f87a1232-993a-476b-919c-d90a6cc42ae8.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Model-based Adversarial Imitation Learning with Self-adatpive Error Control","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"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":"Behavior Cloning, Reinforcement Learning, Adversarial Imitation Learning, Dynamics Models","lastPublishedDoi":"10.21203/rs.3.rs-4263827/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4263827/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Generative Adversarial Imitation Learning (GAIL) presents the ability to learn policies without prior knowledge of the underlying reward function. However, it often suffers from limited sample efficiency due to its reliance on reinforcement learning for policy learning, mandating extensive real-time interactions with the environment. To address this challenge, this paper introduces a refined framework named TM-GAIL, which combines transition function model learning with GAIL. This approach capitalizes on the utility of neural networks to construct a transition function model, facilitating the generation of virtual samples to complement real data. The training of the discriminator is augmented by the inclusion of virtual samples alongside expert demonstration data. In the context of policy learning, the incorporation of virtual samples, real samples, and the reward derived from the discriminator enriches the policy learning. Furthermore, a self-adaptive error control module has been meticulously designed for the regions characterized by high returns and to mitigate model errors. Empirical findings demonstrate that TM-GAIL significantly improves sample efficiency in comparison to imitation learning and model-free methods. It achieves performance levels that closely align with those of domain experts across both continuous and discrete tasks.","manuscriptTitle":"Model-based Adversarial Imitation Learning with Self-adatpive Error Control","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-19 02:46:13","doi":"10.21203/rs.3.rs-4263827/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":"182a546b-7139-4927-9b12-ee8fa3f35c08","owner":[],"postedDate":"April 19th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-06-19T08:24:05+00:00","versionOfRecord":[],"versionCreatedAt":"2024-04-19 02:46:13","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4263827","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4263827","identity":"rs-4263827","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}