{"id":"https://openalex.org/W7160852816","doi":"https://doi.org/10.48550/arxiv.2605.07104","title":"Almost Sure Convergence Rates of Stochastic Approximation and Reinforcement Learning via a Poisson-Moreau Drift","display_name":"Almost Sure Convergence Rates of Stochastic Approximation and Reinforcement Learning via a Poisson-Moreau Drift","publication_year":2026,"publication_date":"2026-05-08","ids":{"openalex":"https://openalex.org/W7160852816","doi":"https://doi.org/10.48550/arxiv.2605.07104"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2605.07104","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2605.07104","pdf_url":"https://arxiv.org/pdf/2605.07104","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":"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/2605.07104","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135848873","display_name":"Xinyu Liu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Liu, Xinyu","raw_affiliation_strings":["University of Virginia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Virginia","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059984944","display_name":"Zixuan Xie","orcid":"https://orcid.org/0000-0002-4921-2309","observed_orcids":["https://orcid.org/0000-0002-4921-2309"]},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xie, Zixuan","raw_affiliation_strings":["University of Virginia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Virginia","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5033834190","display_name":"Shangtong Zhang","orcid":"https://orcid.org/0009-0003-0710-147X","observed_orcids":["https://orcid.org/0009-0003-0710-147X"]},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhang, Shangtong","raw_affiliation_strings":["University of Virginia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Virginia","institution_ids":["https://openalex.org/I51556381"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I51556381"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"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":0.9535999894142151,"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":0.9535999894142151,"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/T12794","display_name":"Adaptive Dynamic Programming Control","score":0.021400000900030136,"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"}},{"id":"https://openalex.org/T11195","display_name":"Simulation Techniques and Applications","score":0.006399999838322401,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/stochastic-approximation","display_name":"stochastic approximation","score":0.996999979019165},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"reinforcement learning","score":0.9929999709129333},{"id":"https://openalex.org/keywords/markovian-noise","display_name":"Markovian noise","score":0.8920000195503235},{"id":"https://openalex.org/keywords/contraction-mapping-principle","display_name":"contractive mappings","score":0.7990000247955322},{"id":"https://openalex.org/keywords/q-learning","display_name":"Q-learning","score":0.9079999923706055},{"id":"https://openalex.org/keywords/lyapunov-drift","display_name":"Lyapunov drift","score":0.9330000281333923},{"id":"https://openalex.org/keywords/poisson-equation","display_name":"Poisson equation","score":0.8080000281333923},{"id":"https://openalex.org/keywords/moreau-envelope","display_name":"Moreau envelope","score":0.8500000238418579},{"id":"https://openalex.org/keywords/law-of-the-iterated-logarithm","display_name":"law of the iterated logarithm","score":0.9539999961853027}],"concepts":[{"id":"https://openalex.org/C55479107","wikidata":"https://www.wikidata.org/wiki/Q97663916","display_name":"Stochastic approximation","level":3,"score":0.6880999803543091},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.6866000294685364},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.6597999930381775},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5964000225067139},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.5008000135421753},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.49779999256134033},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.4884999990463257},{"id":"https://openalex.org/C139907963","wikidata":"https://www.wikidata.org/wiki/Q198740","display_name":"Law of the iterated logarithm","level":3,"score":0.4810999929904938},{"id":"https://openalex.org/C140479938","wikidata":"https://www.wikidata.org/wiki/Q5254619","display_name":"Iterated function","level":2,"score":0.47450000047683716},{"id":"https://openalex.org/C39927690","wikidata":"https://www.wikidata.org/wiki/Q11197","display_name":"Logarithm","level":2,"score":0.46369999647140503},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.44920000433921814},{"id":"https://openalex.org/C8272713","wikidata":"https://www.wikidata.org/wiki/Q176737","display_name":"Stochastic process","level":2,"score":0.4348999857902527},{"id":"https://openalex.org/C3770464","wikidata":"https://www.wikidata.org/wiki/Q775963","display_name":"Smoothing","level":2,"score":0.42579999566078186},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.40529999136924744},{"id":"https://openalex.org/C95763700","wikidata":"https://www.wikidata.org/wiki/Q578985","display_name":"Convergence of random variables","level":3,"score":0.3856000006198883},{"id":"https://openalex.org/C91873725","wikidata":"https://www.wikidata.org/wiki/Q3445816","display_name":"Function approximation","level":3,"score":0.32350000739097595},{"id":"https://openalex.org/C60640748","wikidata":"https://www.wikidata.org/wiki/Q2337858","display_name":"Lyapunov function","level":3,"score":0.311599999666214},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.29829999804496765},{"id":"https://openalex.org/C50479213","wikidata":"https://www.wikidata.org/wiki/Q7617815","display_name":"Stochastic drift","level":2,"score":0.28630000352859497},{"id":"https://openalex.org/C170131372","wikidata":"https://www.wikidata.org/wiki/Q7617811","display_name":"Stochastic control","level":3,"score":0.28200000524520874},{"id":"https://openalex.org/C127934551","wikidata":"https://www.wikidata.org/wiki/Q1148098","display_name":"Harmonic","level":2,"score":0.27720001339912415},{"id":"https://openalex.org/C148764684","wikidata":"https://www.wikidata.org/wiki/Q621751","display_name":"Approximation algorithm","level":2,"score":0.275299996137619},{"id":"https://openalex.org/C188116033","wikidata":"https://www.wikidata.org/wiki/Q2664563","display_name":"Q-learning","level":3,"score":0.274399995803833},{"id":"https://openalex.org/C196340769","wikidata":"https://www.wikidata.org/wiki/Q7698910","display_name":"Temporal difference learning","level":3,"score":0.2653999924659729},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.26350000500679016},{"id":"https://openalex.org/C75438885","wikidata":"https://www.wikidata.org/wiki/Q3403615","display_name":"Large deviations theory","level":2,"score":0.2614000141620636},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.25859999656677246},{"id":"https://openalex.org/C57945734","wikidata":"https://www.wikidata.org/wiki/Q7977941","display_name":"Weak convergence","level":3,"score":0.25}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2605.07104","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2605.07104","pdf_url":"https://arxiv.org/pdf/2605.07104","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2605.07104","doi":"https://doi.org/10.48550/arxiv.2605.07104","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.07104","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:2605.07104","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2605.07104","pdf_url":"https://arxiv.org/pdf/2605.07104","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G2940155921","display_name":"SLES: CRASH - Challenging Reinforcement-learning based Adversarial scenarios for Safety Hardening","funder_award_id":"2331904","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4496510830","display_name":"CAREER: Revolutionizing the Evaluation of AI Agents with Online and Offline Data","funder_award_id":"2442098","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G6671297155","display_name":null,"funder_award_id":"CAREER","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8240942421","display_name":"III: Small: Moving offline learning to rank online, from theory to practice","funder_award_id":"2128019","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320307791","display_name":"Cisco Systems","ror":"https://ror.org/03yt1ez60"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7160852816.pdf","grobid_xml":"https://content.openalex.org/works/W7160852816.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4404650081","https://openalex.org/W4415053945","https://openalex.org/W4390962455","https://openalex.org/W3107656456","https://openalex.org/W7138343762","https://openalex.org/W6891923916","https://openalex.org/W2078618768","https://openalex.org/W3173526354","https://openalex.org/W7084149758","https://openalex.org/W7163010757"],"abstract_inverted_index":{"Establishing":[0],"almost":[1,68,85],"sure":[2,69,86],"convergence":[3,70],"rates":[4],"for":[5,25,55,91,136,144],"stochastic":[6,29],"approximation":[7,30],"and":[8,49],"reinforcement":[9,43],"learning":[10,44,58,81],"under":[11],"Markovian":[12,137],"noise":[13,138],"is":[14,89,96,124],"a":[15,26,37,56,125,132],"fundamental":[16],"theoretical":[17],"challenge.":[18],"We":[19],"make":[20],"progress":[21],"towards":[22],"this":[23],"challenge":[24],"class":[27],"of":[28,116],"algorithms":[31,45],"whose":[32],"expected":[33],"updates":[34],"are":[35,106],"contractive,":[36],"setting":[38],"that":[39,130],"arises":[40],"in":[41],"many":[42],"such":[46],"as":[47],"$Q$-learning":[48],"linear":[50],"temporal":[51],"difference":[52],"learning.":[53],"Specifically,":[54],"power-law":[57],"rate":[59,71,82,88],"$O(n^{-\u03b7})$":[60],"with":[61],"$\u03b7\\in":[62],"(1/2,":[63],"1)$,":[64],"we":[65],"obtain":[66],"an":[67],"arbitrarily":[72,102],"close":[73,103],"to":[74,104,121,139],"$o(n^{1":[75],"-":[76],"2\u03b7})$.":[77],"For":[78],"the":[79,84,97,112,140,145],"harmonic":[80],"$\u03b1_n=\u03b1/(n+1)$,":[83],"squared-error":[87],"$o(n^{-\u03b6})$":[90],"every":[92],"$\u03b6<\\min\\{1,2(1-\u03ba)\u03b1\\}$,":[93],"where":[94],"$\u03ba$":[95],"contraction":[98],"factor.":[99],"Thus":[100],"exponents":[101],"one":[105],"available":[107],"when":[108],"$2(1-\u03ba)\u03b1\\geq1$.":[109],"This":[110],"approaches":[111],"$n^{-1}$":[113],"polynomial":[114],"exponent":[115],"classical":[117],"law-of-the-iterated-logarithm":[118],"benchmarks.":[119],"Key":[120],"our":[122],"analysis":[123],"novel":[126],"Lyapunov":[127],"drift":[128],"construction":[129],"applies":[131],"Poisson-equation":[133],"based":[134],"correction":[135],"well-established":[141],"Moreau-envelope":[142],"smoothing":[143],"contractive":[146],"mapping.":[147]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-05-12T00:00:00"}