{"id":"https://openalex.org/W7220675654","doi":"https://doi.org/10.48550/arxiv.2610.04638","title":"Policy as Data: Replay-Based Policy Dual Averaging via Advantage Regression","display_name":"Policy as Data: Replay-Based Policy Dual Averaging via Advantage Regression","publication_year":2026,"publication_date":"2026-10-03","ids":{"openalex":"https://openalex.org/W7220675654","doi":"https://doi.org/10.48550/arxiv.2610.04638"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.04638","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.04638","pdf_url":"https://arxiv.org/pdf/2610.04638","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/2610.04638","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5015846945","display_name":"Nianli Peng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Peng, Nianli","raw_affiliation_strings":["Harvard University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University ,","institution_ids":["https://openalex.org/I136199984"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012830032","display_name":"Geoffrey J. Gordon","orcid":"https://orcid.org/0009-0007-9960-2114","observed_orcids":["https://orcid.org/0009-0007-9960-2114"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gordon, Geoffrey J.","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5022968100","display_name":"Kiant\u00e9 Brantley","orcid":"https://orcid.org/0000-0002-8395-594X","observed_orcids":["https://orcid.org/0000-0002-8395-594X"]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Brantley, Kiant\u00e9","raw_affiliation_strings":["Harvard University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University ,","institution_ids":["https://openalex.org/I136199984"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":1.0,"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":1.0,"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":0.0,"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":0.0,"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/dual-averaging","display_name":"dual averaging","score":0.5109999775886536},{"id":"https://openalex.org/keywords/actor-critic","display_name":"actor-critic","score":0.7509999871253967},{"id":"https://openalex.org/keywords/sample-efficiency","display_name":"sample efficiency","score":0.5799999833106995},{"id":"https://openalex.org/keywords/mujoco","display_name":"MuJoCo","score":0.6480000019073486},{"id":"https://openalex.org/keywords/atari-games","display_name":"Atari games","score":0.8059999942779541}],"concepts":[{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.7016000151634216},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.5393999814987183},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5120000243186951},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.48660001158714294},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.45509999990463257},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.44940000772476196},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.4300000071525574},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.41110000014305115},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.3919999897480011},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38119998574256897},{"id":"https://openalex.org/C148043351","wikidata":"https://www.wikidata.org/wiki/Q4456944","display_name":"Current (fluid)","level":2,"score":0.37700000405311584},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.37130001187324524},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3659000098705292},{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.34119999408721924},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.31200000643730164},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.30720001459121704},{"id":"https://openalex.org/C152877465","wikidata":"https://www.wikidata.org/wiki/Q208042","display_name":"Regression analysis","level":2,"score":0.30630001425743103},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3043999969959259},{"id":"https://openalex.org/C2780440489","wikidata":"https://www.wikidata.org/wiki/Q5227278","display_name":"Data-driven","level":2,"score":0.28380000591278076},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.28200000524520874},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.26589998602867126}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.04638","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.04638","pdf_url":"https://arxiv.org/pdf/2610.04638","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.2610.04638","doi":"https://doi.org/10.48550/arxiv.2610.04638","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.04638","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:2610.04638","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.04638","pdf_url":"https://arxiv.org/pdf/2610.04638","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/W7220675654.pdf","grobid_xml":"https://content.openalex.org/works/W7220675654.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7084098751","https://openalex.org/W2948708918","https://openalex.org/W3207882166","https://openalex.org/W3032956793","https://openalex.org/W2089434629","https://openalex.org/W4281250801","https://openalex.org/W3204075372","https://openalex.org/W7167912155","https://openalex.org/W2556958149","https://openalex.org/W2787613197"],"abstract_inverted_index":{"Actor-critic":[0],"methods":[1],"reuse":[2],"past":[3],"experience":[4],"to":[5,37,58,65,84,135],"improve":[6],"sample":[7],"efficiency.":[8],"However,":[9],"historical":[10,53],"data":[11],"are":[12],"typically":[13],"regarded":[14],"as":[15],"off-policy":[16],"samples":[17,73],"for":[18],"the":[19,43,59,66,85,90,94,99,114,128,148,185],"current":[20,91],"policy-improvement":[21,50],"update.":[22,69],"This":[23],"work":[24],"introduces":[25],"Regularized":[26],"Dual":[27],"Averaging":[28],"Actor":[29],"Critic":[30],"(RDA2C),":[31],"which":[32,113,151],"assigns":[33],"a":[34,79,123],"distinct":[35],"role":[36],"replay.":[38],"In":[39,103],"regularized":[40,129],"dual":[41,80,110],"averaging,":[42],"subsequent":[44],"policy":[45,92,115],"is":[46,116,154],"determined":[47],"by":[48],"accumulated":[49,95],"feedback,":[51],"so":[52],"advantage":[54,76,158],"estimates":[55],"contribute":[56],"directly":[57],"actor":[60],"objective":[61,111],"rather":[62],"than":[63],"solely":[64],"most":[67],"recent":[68],"RDA2C":[70,156,166,181,198],"stores":[71],"state-action":[72],"with":[74],"critic-estimated":[75],"labels,":[77,165,197],"fits":[78],"score":[81,96],"model":[82,97],"$Z_\u03b8$":[83],"aggregated":[86],"dataset,":[87],"and":[88,143,146,175,205],"derives":[89],"from":[93,112,132,160],"using":[98],"entropy":[100],"mirror":[101],"map.":[102],"this":[104],"way,":[105],"replay":[106,144],"defines":[107],"an":[108],"empirical":[109],"computed.":[117],"To":[118],"analyze":[119],"RDA2C,":[120],"we":[121],"establish":[122],"finite-time":[124],"value-gap":[125],"decomposition,":[126],"separating":[127],"dual-averaging":[130,187],"term":[131],"errors":[133],"due":[134],"stale-replay":[136],"supervised":[137],"fitting,":[138],"critic":[139],"bias,":[140],"finite-buffer":[141],"variance,":[142],"coverage,":[145],"stating":[147],"assumptions":[149],"under":[150],"each":[152],"error":[153],"bounded.":[155],"accepts":[157],"labels":[159],"any":[161],"critic.":[162],"With":[163,195],"GAE":[164],"outperforms":[167,183],"PPO":[168],"on":[169,189],"six":[170,190],"of":[171,177,191],"eight":[172,176,192],"MuJoCo":[173,193],"tasks":[174],"twelve":[178],"Atari":[179],"games.":[180],"also":[182],"AAPDA,":[184],"closest":[186],"baseline,":[188],"tasks.":[194],"twin-$Q$":[196],"matches":[199],"SAC":[200],"at":[201],"matched":[202],"batch":[203],"size":[204],"update":[206],"frequency.":[207]},"counts_by_year":[],"updated_date":"2026-11-06T12:05:46.655746","created_date":"2026-10-07T00:00:00"}