{"id":"https://openalex.org/W7221802804","doi":"https://doi.org/10.48550/arxiv.2610.14988","title":"Mixing Regularization for Average-Reward $Q$-Learning and $Q$-Planning","display_name":"Mixing Regularization for Average-Reward $Q$-Learning and $Q$-Planning","publication_year":2026,"publication_date":"2026-10-10","ids":{"openalex":"https://openalex.org/W7221802804","doi":"https://doi.org/10.48550/arxiv.2610.14988"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.14988","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.14988","pdf_url":"https://arxiv.org/pdf/2610.14988","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.14988","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136259016","display_name":"Julien Roger","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I142476485","display_name":"\u00c9cole Polytechnique","ror":"https://ror.org/05hy3tk52","country_code":"FR","type":"education","lineage":["https://openalex.org/I142476485","https://openalex.org/I4210145102"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Roger, Julien","raw_affiliation_strings":["\u00c9cole polytechnique"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"\u00c9cole polytechnique","institution_ids":["https://openalex.org/I142476485"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119738332","display_name":"Sam Bodmer","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bodmer, Sam","raw_affiliation_strings":["New York University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"New York University","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108813500","display_name":"Jian Qian","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Qian, Jian","raw_affiliation_strings":["The University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5087653321","display_name":"Benjamin Rivi\u00e8re","orcid":"https://orcid.org/0000-0003-4189-4090","observed_orcids":["https://orcid.org/0000-0003-4189-4090","https://orcid.org/0000-0002-0597-5400"]},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Riviere, Benjamin","raw_affiliation_strings":["New York University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"New York University","institution_ids":["https://openalex.org/I57206974"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9994999766349792,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9994999766349792,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12101","display_name":"Advanced Bandit Algorithms Research","score":9.999999747378752e-05,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12794","display_name":"Adaptive Dynamic Programming Control","score":9.999999747378752e-05,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/average-reward-reinforcement-learning","display_name":"average-reward reinforcement learning","score":0.9710000157356262},{"id":"https://openalex.org/keywords/q-learning","display_name":"Q-learning","score":0.8539999723434448},{"id":"https://openalex.org/keywords/contraction-analysis","display_name":"contraction analysis","score":0.47200000286102295},{"id":"https://openalex.org/keywords/finite-time-analysis","display_name":"finite-time analysis","score":0.6019999980926514},{"id":"https://openalex.org/keywords/exploration","display_name":"exploration","score":0.5139999985694885},{"id":"https://openalex.org/keywords/operator-bias","display_name":"operator bias","score":0.6629999876022339},{"id":"https://openalex.org/keywords/alpha-divergence","display_name":"alpha-divergence","score":0.6679999828338623},{"id":"https://openalex.org/keywords/sample-complexity","display_name":"sample complexity","score":0.6600000262260437},{"id":"https://openalex.org/keywords/planning-horizon","display_name":"planning horizon","score":0.5460000038146973},{"id":"https://openalex.org/keywords/markov-chain-mixing","display_name":"Markov chain mixing","score":0.7089999914169312}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6463000178337097},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6392999887466431},{"id":"https://openalex.org/C138777275","wikidata":"https://www.wikidata.org/wiki/Q6884054","display_name":"Mixing (physics)","level":2,"score":0.6074000000953674},{"id":"https://openalex.org/C17020691","wikidata":"https://www.wikidata.org/wiki/Q139677","display_name":"Operator (biology)","level":5,"score":0.48010000586509705},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.47839999198913574},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.4627000093460083},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.45980000495910645},{"id":"https://openalex.org/C28761237","wikidata":"https://www.wikidata.org/wiki/Q7805321","display_name":"Time horizon","level":2,"score":0.4530999958515167},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.44839999079704285},{"id":"https://openalex.org/C163415756","wikidata":"https://www.wikidata.org/wiki/Q126473","display_name":"Contraction (grammar)","level":2,"score":0.44839999079704285},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.4325000047683716},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.43149998784065247},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.40709999203681946},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.38839998841285706},{"id":"https://openalex.org/C91575142","wikidata":"https://www.wikidata.org/wiki/Q1971426","display_name":"Optimal control","level":2,"score":0.36410000920295715},{"id":"https://openalex.org/C188116033","wikidata":"https://www.wikidata.org/wiki/Q2664563","display_name":"Q-learning","level":3,"score":0.362199991941452},{"id":"https://openalex.org/C182310444","wikidata":"https://www.wikidata.org/wiki/Q1332643","display_name":"Boundary value problem","level":2,"score":0.3596999943256378},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.3400000035762787},{"id":"https://openalex.org/C2780695315","wikidata":"https://www.wikidata.org/wiki/Q3799040","display_name":"Unobservable","level":2,"score":0.33570000529289246},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.2928999960422516},{"id":"https://openalex.org/C200741047","wikidata":"https://www.wikidata.org/wiki/Q1957758","display_name":"Integrable system","level":2,"score":0.29019999504089355},{"id":"https://openalex.org/C61445026","wikidata":"https://www.wikidata.org/wiki/Q217608","display_name":"Fixed point","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.2806999981403351},{"id":"https://openalex.org/C144237770","wikidata":"https://www.wikidata.org/wiki/Q747534","display_name":"Mathematical economics","level":1,"score":0.2678000032901764},{"id":"https://openalex.org/C38858127","wikidata":"https://www.wikidata.org/wiki/Q5441228","display_name":"Feed forward","level":2,"score":0.26499998569488525},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.26429998874664307},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2606000006198883},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.2605000138282776},{"id":"https://openalex.org/C14646407","wikidata":"https://www.wikidata.org/wiki/Q1430750","display_name":"Bellman equation","level":2,"score":0.25760000944137573}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.14988","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.14988","pdf_url":"https://arxiv.org/pdf/2610.14988","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2610.14988","doi":"https://doi.org/10.48550/arxiv.2610.14988","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.14988","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.14988","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.14988","pdf_url":"https://arxiv.org/pdf/2610.14988","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7221802804.pdf","grobid_xml":"https://content.openalex.org/works/W7221802804.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7123357605","https://openalex.org/W2160284799","https://openalex.org/W4226128862","https://openalex.org/W7127657275","https://openalex.org/W4406060230","https://openalex.org/W2945884844","https://openalex.org/W7126264174","https://openalex.org/W2460675832","https://openalex.org/W6884998335","https://openalex.org/W3200702333"],"abstract_inverted_index":{"The":[0],"average-reward":[1],"setting":[2,41],"for":[3,119,125],"reinforcement":[4],"learning":[5],"is":[6,12,24],"challenging":[7],"because":[8],"its":[9,182],"Bellman":[10],"operator":[11,106],"known":[13,117],"to":[14,90],"contract":[15],"in":[16],"span":[17],"only":[18],"when":[19],"the":[20,40,54,60,66,91,109,115,146,152,155,159,168,193,201],"underlying":[21],"Markov":[22],"chain":[23],"mixing.":[25],"Whereas":[26],"existing":[27],"work":[28],"typically":[29],"assumes":[30],"a":[31,46,173],"contraction":[32,55],"property,":[33],"e.g.,":[34],"through":[35],"environment":[36],"mixing,":[37],"we":[38,43,52,128,163],"consider":[39],"where":[42,145],"are":[44],"given":[45],"reward-agnostic":[47],"$m$-step":[48,126],"mixing":[49,67,127,175],"policy,":[50],"and":[51,56,105,121,124,137,179,181,195],"control":[53,102],"exploration":[57],"properties":[58],"of":[59,73,83,172,198],"learned":[61],"policy":[62,176],"by":[63],"regularizing":[64],"toward":[65],"policy.":[68],"We":[69,132],"provide":[70],"finite-time":[71],"analysis":[72],"$Q$-learning":[74],"algorithms":[75],"that":[76,101,139],"can":[77],"be":[78],"instantiated":[79],"with":[80,85,88,185],"multiple":[81],"choices":[82],"regularization,":[84],"guarantees":[86],"measured":[87],"respect":[89],"regularized":[92],"fixed":[93],"point.":[94],"Our":[95],"framework":[96],"exposes":[97],"novel":[98],"regularizer-dependent":[99],"objects":[100],"contraction,":[103],"exploration,":[104,180],"bias.":[107],"In":[108],"one-step":[110],"setting,":[111],"our":[112,190],"rates":[113,118],"match":[114],"best":[116],"$Q$-iteration":[120],"synchronous":[122],"$Q$-learning,":[123],"derive":[129],"new":[130],"rates.":[131],"also":[133],"study":[134],"$Q$-planning":[135],"variants,":[136],"show":[138],"planning":[140,147],"acts":[141],"as":[142],"time-localized":[143],"learning,":[144],"horizon":[148],"$H$":[149],"trades":[150],"off":[151],"bias":[153,196],"from":[154,200],"boundary":[156],"value":[157],"against":[158],"sample":[160],"complexity.":[161],"Finally,":[162],"present":[164],"numerical":[165],"studies":[166],"validating":[167],"analysis:":[169],"(i)":[170],"knowledge":[171],"strong":[174],"improves":[177],"convergence":[178,194],"advantage":[183],"increases":[184],"bottleneck":[186],"severity":[187],"and,":[188],"(ii)":[189],"theory":[191],"predicts":[192],"rankings":[197],"regularizers":[199],"$\u03b1$-divergence":[202],"family.":[203]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-10-15T00:00:00"}