{"id":"https://openalex.org/W7160961394","doi":"https://doi.org/10.48550/arxiv.2605.08253","title":"Path-Coupled Bellman Flows for Distributional Reinforcement Learning","display_name":"Path-Coupled Bellman Flows for Distributional Reinforcement Learning","publication_year":2026,"publication_date":"2026-05-07","ids":{"openalex":"https://openalex.org/W7160961394","doi":"https://doi.org/10.48550/arxiv.2605.08253"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.08253","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08253","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.08253","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5011889355","display_name":"Boyang Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Boyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135968829","display_name":"Qing Zou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zou, Qing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135920499","display_name":"Siqin Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Siqin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135977234","display_name":"Hao Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Hao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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":0.6488999724388123,"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":0.6488999724388123,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.03909999877214432,"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/T10784","display_name":"Muscle activation and electromyography studies","score":0.03759999945759773,"subfield":{"id":"https://openalex.org/subfields/2204","display_name":"Biomedical 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/successor-cardinal","display_name":"Successor cardinal","score":0.6362000107765198},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5845000147819519},{"id":"https://openalex.org/keywords/base","display_name":"Base (topology)","score":0.4487000107765198},{"id":"https://openalex.org/keywords/current","display_name":"Current (fluid)","score":0.41449999809265137},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.40880000591278076},{"id":"https://openalex.org/keywords/bellman-equation","display_name":"Bellman equation","score":0.39100000262260437},{"id":"https://openalex.org/keywords/variance","display_name":"Variance (accounting)","score":0.35929998755455017}],"concepts":[{"id":"https://openalex.org/C75306776","wikidata":"https://www.wikidata.org/wiki/Q7632662","display_name":"Successor cardinal","level":2,"score":0.6362000107765198},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5845000147819519},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5461999773979187},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.44999998807907104},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.4487000107765198},{"id":"https://openalex.org/C148043351","wikidata":"https://www.wikidata.org/wiki/Q4456944","display_name":"Current (fluid)","level":2,"score":0.41449999809265137},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.40880000591278076},{"id":"https://openalex.org/C14646407","wikidata":"https://www.wikidata.org/wiki/Q1430750","display_name":"Bellman equation","level":2,"score":0.39100000262260437},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.35929998755455017},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.35260000824928284},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.3522999882698059},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3458000123500824},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.3310000002384186},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3140999972820282},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.2969000041484833},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.29170000553131104},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.27489998936653137},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.27160000801086426},{"id":"https://openalex.org/C92757383","wikidata":"https://www.wikidata.org/wiki/Q382497","display_name":"Affine transformation","level":2,"score":0.25609999895095825}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.08253","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08253","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.08253","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08253","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Distributional":[0],"reinforcement":[1],"learning":[2],"(DRL)":[3],"models":[4],"the":[5,28,63,68,75,87],"full":[6],"return":[7,54,112],"distribution,":[8],"but":[9],"existing":[10],"finite-support":[11],"or":[12,31],"quantile-based":[13],"methods":[14],"rely":[15],"on":[16,140],"projections,":[17],"while":[18,131],"recent":[19],"flow-based":[20],"approaches":[21],"can":[22],"suffer":[23],"from":[24,32,67],"\\emph{boundary":[25],"mismatch}":[26],"at":[27,72,78,90],"flow":[29,57,89],"source":[30],"\\emph{high-variance}":[33],"bootstrapping":[34],"when":[35],"current":[36,64,109],"and":[37,80,110,118,145,151,154],"successor":[38,88,111],"noises":[39],"are":[40],"independent.":[41],"We":[42],"propose":[43],"Path-Coupled":[44],"Bellman":[45,76,101,129],"Flows":[46],"(PCBF),":[47],"a":[48,82,99,120],"continuous-time":[49],"DRL":[50],"method":[51],"that":[52],"learns":[53],"distributions":[55],"with":[56],"matching":[58],"using":[59],"\\textbf{source-consistent":[60],"Bellman-coupled":[61],"paths}:":[62],"path":[65],"starts":[66],"required":[69],"base":[70,116],"prior":[71],"$t{=}0$,":[73],"reaches":[74],"target":[77],"$t{=}1$,":[79],"maintains":[81],"pathwise":[83],"affine":[84],"relation":[85],"to":[86,97],"intermediate":[91],"times":[92],"(without":[93],"requiring":[94],"time-$t$":[95],"marginals":[96],"satisfy":[98],"distributional":[100,149],"fixed":[102],"point":[103],"for":[104,136],"all":[105],"$t$).":[106],"PCBF":[107],"couples":[108],"flows":[113],"through":[114],"shared":[115],"noise":[117],"uses":[119],"$\u03bb$-parameterized":[121],"control-variate":[122],"target:":[123],"$\u03bb{=}0$":[124],"recovers":[125],"an":[126],"unbiased":[127],"sample":[128],"target,":[130],"$\u03bb{>}0$":[132],"trades":[133],"controlled":[134],"bias":[135],"variance":[137],"reduction.":[138],"Experiments":[139],"analytically":[141],"tractable":[142],"MRPs,":[143],"OGBench,":[144],"D4RL":[146],"show":[147],"improved":[148],"fidelity":[150],"training":[152],"stability,":[153],"competitive":[155],"offline":[156],"RL":[157],"performance.":[158]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-13T00:00:00"}