Successively Pruned Q-Learning: Using Self Q-function to Reduce the Overestimation

Successively Pruned Q-Learning: Using Self Q-function to Reduce the Overestimation | 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 Successively Pruned Q-Learning: Using Self Q-function to Reduce the Overestimation Li He, Zhaolin Xue, Zhongxue Gan, Lihua Zhang, Zhiyan Dong This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5287151/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 2 You are reading this latest preprint version Abstract It’s well-known that the Q-learning algorithm suffers the overestimation owingto using the maximum state-action value as an approximation of the maximumexpected state-action value. Double Q-learning and other algorithms have beenproposed as efficient solutions to alleviate the overestimation. However, theseproposed methods intend to utilize multiple Q-functions to reduce the overes-timation and ignore the information of single Q-function. In this paper, 1) wereinterpret the update process of Q-learning, build a more precise model compat-ible with previous model. 2) We propose a novel and simple method to controlthe maximum bias by employing the information of single Q-function. 3) Ourmethod not only balances between the overestimation and the underestimation,but also attains the minimum bias under proper hyper-parameters. 4) Moreover,it can be naturally generalized to the discrete control domain and continuouscontrol tasks. We reveal that our algorithms outperform Double DQN and otheralgorithms on some representative games. Additionally, classical off-policy actor-critic algorithms also gain benefits from our method.Ultimately, we have extendedour algorithm to multi-agent reinforcement learning algorithms. Successively pruned Q-learning Overestimation DQN SAC Multi-agent reinforcement learning Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Submission checks completed at journal 19 Oct, 2024 First submitted to journal 18 Oct, 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-5287151","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":367895278,"identity":"03d0345e-bd94-4d7a-ba8b-2c38e7f5cc04","order_by":0,"name":"Li He","email":"","orcid":"","institution":"Fudan University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"He","suffix":""},{"id":367895279,"identity":"97a66aac-880b-4ce9-ade1-172f548ca784","order_by":1,"name":"Zhaolin Xue","email":"","orcid":"","institution":"Fudan University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhaolin","middleName":"","lastName":"Xue","suffix":""},{"id":367895280,"identity":"4d0608d2-6a8d-4079-aa76-cb3a56bc0378","order_by":2,"name":"Zhongxue Gan","email":"","orcid":"","institution":"Fudan University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhongxue","middleName":"","lastName":"Gan","suffix":""},{"id":367895281,"identity":"569ab4da-99ac-4858-9d2d-cc08c7ea6798","order_by":3,"name":"Lihua Zhang","email":"","orcid":"","institution":"Fudan University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Lihua","middleName":"","lastName":"Zhang","suffix":""},{"id":367895282,"identity":"fbe24ba5-07bf-4865-bd0c-301b53330b77","order_by":4,"name":"Zhiyan Dong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAElEQVRIiWNgGAWjYDCCA0DM2CABJJkPfGBgA4sZEKuFLXEGKVpALB5D4rTwHe89/PLnDovE/tk9H5t5yuzyGNibt0kw1NzBqUXyzLk0C8kzEokz7pzd2MxzLrmYgedYmQTDsWc4tRjcyDEzMGyTSGy4kbv9MW8bc2KDRI6ZBGPDYfxaEoFa5t/IedjM21af2CD/hqAW4wcHgVo23MhhBGo5DLSFB78WyTNnzBgb2ySMN95IM2ycc+54YhtPWrFFwjHcWviO9xh//NlWJzvvRvLDhjdl1Yn97Ic33vhQg1sLELCB4tGxAc4FEQn4NAATygcgYY9fzSgYBaNgFIxoAAC2DF2o/XtVpgAAAABJRU5ErkJggg==","orcid":"","institution":"Fudan University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Zhiyan","middleName":"","lastName":"Dong","suffix":""}],"badges":[],"createdAt":"2024-10-18 07:08:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5287151/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5287151/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":67809992,"identity":"cfcc5880-0f2e-44b1-921c-3e1f99ae64b0","added_by":"auto","created_at":"2024-10-30 02:50:30","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":845927,"visible":true,"origin":"","legend":"","description":"","filename":"SuccessivelyPrunedQLearning.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5287151/v1_covered_0b14f145-9101-43e5-a1a6-480e0fb70a6e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Successively Pruned Q-Learning: Using Self Q-function to Reduce the Overestimation","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"autonomous-agents-and-multi-agent-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"agnt","sideBox":"Learn more about [Autonomous Agents and Multi-Agent Systems](http://link.springer.com/journal/10458)","snPcode":"10458","submissionUrl":"https://submission.nature.com/new-submission/10458/3","title":"Autonomous Agents and Multi-Agent Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Successively pruned Q-learning, Overestimation, DQN, SAC, Multi-agent reinforcement learning","lastPublishedDoi":"10.21203/rs.3.rs-5287151/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5287151/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"It’s well-known that the Q-learning algorithm suffers the overestimation owingto using the maximum state-action value as an approximation of the maximumexpected state-action value. Double Q-learning and other algorithms have beenproposed as efficient solutions to alleviate the overestimation. However, theseproposed methods intend to utilize multiple Q-functions to reduce the overes-timation and ignore the information of single Q-function. In this paper, 1) wereinterpret the update process of Q-learning, build a more precise model compat-ible with previous model. 2) We propose a novel and simple method to controlthe maximum bias by employing the information of single Q-function. 3) Ourmethod not only balances between the overestimation and the underestimation,but also attains the minimum bias under proper hyper-parameters. 4) Moreover,it can be naturally generalized to the discrete control domain and continuouscontrol tasks. We reveal that our algorithms outperform Double DQN and otheralgorithms on some representative games. Additionally, classical off-policy actor-critic algorithms also gain benefits from our method.Ultimately, we have extendedour algorithm to multi-agent reinforcement learning algorithms.","manuscriptTitle":"Successively Pruned Q-Learning: Using Self Q-function to Reduce the Overestimation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-10-30 02:42:23","doi":"10.21203/rs.3.rs-5287151/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"checksComplete","content":"","date":"2024-10-19T06:53:06+00:00","index":"","fulltext":""},{"type":"submitted","content":"Autonomous Agents and Multi-Agent Systems","date":"2024-10-18T07:05:21+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"autonomous-agents-and-multi-agent-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"agnt","sideBox":"Learn more about [Autonomous Agents and Multi-Agent Systems](http://link.springer.com/journal/10458)","snPcode":"10458","submissionUrl":"https://submission.nature.com/new-submission/10458/3","title":"Autonomous Agents and Multi-Agent Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"7d5a09de-65cf-479f-a6c4-fa31349755ff","owner":[],"postedDate":"October 30th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-13T03:10:20+00:00","versionOfRecord":[],"versionCreatedAt":"2024-10-30 02:42:23","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5287151","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5287151","identity":"rs-5287151","version":["v1"]},"buildId":"-HB7Z8yhvgn0wM9Nzuekk","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}