{"id":"https://openalex.org/W7214717020","doi":"https://doi.org/10.48550/arxiv.2609.32727","title":"Distributionally Robust Average-Reward Reinforcement Learning: Finite-Sample Guarantees under Weak Communication","display_name":"Distributionally Robust Average-Reward Reinforcement Learning: Finite-Sample Guarantees under Weak Communication","publication_year":2026,"publication_date":"2026-09-26","ids":{"openalex":"https://openalex.org/W7214717020","doi":"https://doi.org/10.48550/arxiv.2609.32727"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.32727","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.32727","pdf_url":"https://arxiv.org/pdf/2609.32727","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","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.32727","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5153412397","display_name":"Chenyu Lu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Lu, Chenyu","raw_affiliation_strings":["The Hong Kong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153523466","display_name":"Zijun Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Chen, Zijun","raw_affiliation_strings":["The Hong Kong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5027388649","display_name":"Nian Si","orcid":"https://orcid.org/0000-0002-4730-543X","observed_orcids":["https://orcid.org/0000-0002-4730-543X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Si, Nian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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":0.9945999979972839,"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":0.9945999979972839,"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":0.002400000113993883,"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":0.0017999999690800905,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"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":0.7680000066757202},{"id":"https://openalex.org/keywords/finite-sample-guarantees","display_name":"finite-sample guarantees","score":0.9710000157356262},{"id":"https://openalex.org/keywords/sample-complexity","display_name":"sample complexity","score":0.8199999928474426},{"id":"https://openalex.org/keywords/kullback-leibler-divergence","display_name":"Kullback-Leibler divergence","score":0.7680000066757202},{"id":"https://openalex.org/keywords/f-divergence","display_name":"f-divergence","score":0.8410000205039978},{"id":"https://openalex.org/keywords/wasserstein-ball","display_name":"Wasserstein ball","score":0.7860000133514404}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6284999847412109},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.6204000115394592},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.6158999800682068},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.5903000235557556},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.5013999938964844},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.4602999985218048},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.4578999876976013},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.4214000105857849},{"id":"https://openalex.org/C193254401","wikidata":"https://www.wikidata.org/wiki/Q2160088","display_name":"Robust optimization","level":2,"score":0.38029998540878296},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.35760000348091125},{"id":"https://openalex.org/C149441793","wikidata":"https://www.wikidata.org/wiki/Q200726","display_name":"Probability distribution","level":2,"score":0.3515999913215637},{"id":"https://openalex.org/C31531917","wikidata":"https://www.wikidata.org/wiki/Q915157","display_name":"Robust control","level":3,"score":0.34549999237060547},{"id":"https://openalex.org/C37404715","wikidata":"https://www.wikidata.org/wiki/Q380679","display_name":"Dynamic programming","level":2,"score":0.3449000120162964},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3357999920845032},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.3273000121116638},{"id":"https://openalex.org/C141042865","wikidata":"https://www.wikidata.org/wiki/Q200125","display_name":"Expected value","level":2,"score":0.3203999996185303},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.31380000710487366},{"id":"https://openalex.org/C14646407","wikidata":"https://www.wikidata.org/wiki/Q1430750","display_name":"Bellman equation","level":2,"score":0.31360000371932983},{"id":"https://openalex.org/C178635117","wikidata":"https://www.wikidata.org/wiki/Q747499","display_name":"RADIUS","level":2,"score":0.303600013256073},{"id":"https://openalex.org/C2778445095","wikidata":"https://www.wikidata.org/wiki/Q18354077","display_name":"Sample complexity","level":2,"score":0.2994000017642975},{"id":"https://openalex.org/C181243257","wikidata":"https://www.wikidata.org/wiki/Q1693522","display_name":"Sample mean and sample covariance","level":3,"score":0.29030001163482666},{"id":"https://openalex.org/C91575142","wikidata":"https://www.wikidata.org/wiki/Q1971426","display_name":"Optimal control","level":2,"score":0.2799000144004822},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.2711000144481659},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.259799987077713},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2556000053882599}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.32727","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.32727","pdf_url":"https://arxiv.org/pdf/2609.32727","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.2609.32727","doi":"https://doi.org/10.48550/arxiv.2609.32727","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.32727","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:2609.32727","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.32727","pdf_url":"https://arxiv.org/pdf/2609.32727","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":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214717020.pdf","grobid_xml":"https://content.openalex.org/works/W7214717020.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4415199079","https://openalex.org/W4323566444","https://openalex.org/W4417300438","https://openalex.org/W7117108595","https://openalex.org/W4414786512","https://openalex.org/W4403816281","https://openalex.org/W4378713459","https://openalex.org/W7202112104","https://openalex.org/W4416186284","https://openalex.org/W4282976760"],"abstract_inverted_index":{"We":[0,129],"study":[1],"distributionally":[2],"robust":[3,23,58,100,125],"reinforcement":[4],"learning":[5,28,108],"(DR-RL)":[6],"in":[7],"the":[8,22,57,79,99,115,142],"average-reward":[9,59],"setting":[10],"under":[11,55],"weak":[12],"communication.":[13],"Our":[14,86],"main":[15],"result":[16],"provides":[17],"finite-sample":[18],"guarantees":[19],"for":[20,45,67,97,107],"estimating":[21,98],"optimal":[24,101,126],"average":[25,102],"reward":[26,103],"and":[27,35,40,47,70,90,104,121],"a":[29,63,124],"near-optimal":[30],"policy,":[31],"covering":[32],"both":[33],"SA-rectangular":[34],"S-rectangular":[36],"structures":[37],"with":[38],"divergence-based":[39],"distance-based":[41],"uncertainty":[42],"sets.":[43],"Specifically,":[44],"Kullback--Leibler":[46],"$f_k$-divergence":[48],"balls,":[49,72],"we":[50,140],"establish":[51],"explicit":[52,134],"radius":[53,75],"conditions":[54],"which":[56],"Bellman":[60],"equation":[61],"admits":[62],"constant-gain":[64],"solution,":[65],"while":[66],"total":[68],"variation":[69],"Wasserstein":[71],"any":[73],"positive":[74,117],"suffices":[76],"without":[77],"requiring":[78],"nominal":[80,118],"MDP":[81],"to":[82],"be":[83],"weakly":[84],"communicating.":[85],"algorithm":[87],"is":[88,114,123],"prior-knowledge-free":[89],"achieves":[91],"sample":[92],"complexities":[93],"of":[94],"$\\widetilde":[95,105],"O(|\\mathcal{S}||\\mathcal{A}|p_{\\wedge}^{-1}\\operatorname{Span}^{2}(u_\u03b4^{\\ast})\u03b5^{-2})$":[96],"O(|\\mathcal{S}||\\mathcal{A}|p_{\\wedge}^{-2}\\operatorname{Span}^{2}(u_\u03b4^{\\ast})\u03b5^{-2})$":[106],"an":[109,132],"$\u03b5$-optimal":[110],"policy.":[111],"Here,":[112],"$p_{\\wedge}$":[113],"smallest":[116],"transition":[119],"probability":[120],"$u_\u03b4^{\\ast}$":[122],"bias":[127],"function.":[128],"further":[130],"provide":[131],"almost-tight":[133],"upper":[135],"bound":[136],"on":[137],"$\\operatorname{Span}(u_\u03b4^{\\ast})$.":[138],"Finally,":[139],"validate":[141],"predicted":[143],"$n^{-1/2}$":[144],"convergence":[145],"rate":[146],"through":[147],"numerical":[148],"experiments.":[149]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-09-30T00:00:00"}