{"id":"https://openalex.org/W2990210896","doi":"https://doi.org/10.48550/arxiv.1906.06321","title":"Provably Efficient $Q$-learning with Function Approximation via Distribution Shift Error Checking Oracle","display_name":"Provably Efficient $Q$-learning with Function Approximation via Distribution Shift Error Checking Oracle","publication_year":2019,"publication_date":"2019-06-14","ids":{"openalex":"https://openalex.org/W2990210896","doi":"https://doi.org/10.48550/arxiv.1906.06321","mag":"2990210896"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:1906.06321","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1906.06321","pdf_url":"https://arxiv.org/pdf/1906.06321","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/1906.06321","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5033061754","display_name":"Simon S. Du","orcid":"https://orcid.org/0000-0003-0056-8299","observed_orcids":["https://orcid.org/0000-0003-0056-8299"]},"institutions":[{"id":"https://openalex.org/I40036882","display_name":"Institute for Advanced Study","ror":"https://ror.org/00f809463","country_code":"US","type":"facility","lineage":["https://openalex.org/I40036882"]},{"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":"Du, Simon S.","raw_affiliation_strings":["Institute for Advanced Study ,","Carnegie Mellon University, Pittsburgh, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute for Advanced Study ,","institution_ids":["https://openalex.org/I40036882"]},{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, United States","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102363880","display_name":"Yuping Luo","orcid":"https://orcid.org/0009-0002-8483-395X","observed_orcids":["https://orcid.org/0009-0002-8483-395X"]},"institutions":[{"id":"https://openalex.org/I20089843","display_name":"Princeton University","ror":"https://ror.org/00hx57361","country_code":"US","type":"education","lineage":["https://openalex.org/I20089843"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Luo, Yuping","raw_affiliation_strings":["Princeton University ,","Princeton University, Princeton, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Princeton University ,","institution_ids":["https://openalex.org/I20089843"]},{"raw_affiliation_string":"Princeton University, Princeton, United States","institution_ids":["https://openalex.org/I20089843"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062744744","display_name":"Ruosong Wang","orcid":"https://orcid.org/0000-0002-2148-0993","observed_orcids":["https://orcid.org/0000-0002-2148-0993"]},"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":"Wang, Ruosong","raw_affiliation_strings":["Carnegie Mellon University ,","Carnegie Mellon University, Pittsburgh, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University ,","institution_ids":["https://openalex.org/I74973139"]},{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, United States","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5091050078","display_name":"Hanrui Zhang","orcid":"https://orcid.org/0000-0001-9924-0319","observed_orcids":["https://orcid.org/0000-0001-9924-0319"]},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhang, Hanrui","raw_affiliation_strings":["Duke University","Duke University, Durham, United States"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Duke University","institution_ids":["https://openalex.org/I170897317"]},{"raw_affiliation_string":"Duke University, Durham, United States","institution_ids":["https://openalex.org/I170897317"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":48,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":"32","issue":null,"first_page":"8058","last_page":"8068"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9990000128746033,"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":0.9990000128746033,"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":0.000699999975040555,"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/T12072","display_name":"Machine Learning and Algorithms","score":9.999999747378752e-05,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/q-learning","display_name":"Q-learning","score":0.9620000123977661},{"id":"https://openalex.org/keywords/function-approximation","display_name":"function approximation","score":0.9070000052452087},{"id":"https://openalex.org/keywords/linear-function-approximation","display_name":"linear function approximation","score":0.9419999718666077},{"id":"https://openalex.org/keywords/exploration","display_name":"exploration","score":0.382999986410141},{"id":"https://openalex.org/keywords/near-optimal-policies","display_name":"near-optimal policies","score":0.753000020980835}],"concepts":[{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.843299150466919},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6935513615608215},{"id":"https://openalex.org/C91873725","wikidata":"https://www.wikidata.org/wiki/Q3445816","display_name":"Function approximation","level":3,"score":0.6830863356590271},{"id":"https://openalex.org/C148764684","wikidata":"https://www.wikidata.org/wiki/Q621751","display_name":"Approximation algorithm","level":2,"score":0.6028957366943359},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.5765578150749207},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4616507291793823},{"id":"https://openalex.org/C188116033","wikidata":"https://www.wikidata.org/wiki/Q2664563","display_name":"Q-learning","level":3,"score":0.4515973925590515},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.39755141735076904},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3973383605480194},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.3590717613697052},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.35685214400291443},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.19691941142082214},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.13102832436561584},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C78458016","wikidata":"https://www.wikidata.org/wiki/Q840400","display_name":"Evolutionary biology","level":1,"score":0.0},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:1906.06321","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1906.06321","pdf_url":"https://arxiv.org/pdf/1906.06321","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.1906.06321","doi":"https://doi.org/10.48550/arxiv.1906.06321","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1906.06321","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},{"id":"mag:2990210896","doi":null,"is_oa":false,"landing_page_url":"https://papers.nips.cc/paper/2019/file/3d8e03e8b133b16f13a586f0c01b6866-Paper.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":"Neural Information Processing Systems","raw_type":null}],"best_oa_location":{"id":"pmh:oai:arXiv.org:1906.06321","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1906.06321","pdf_url":"https://arxiv.org/pdf/1906.06321","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":[{"id":"https://openalex.org/F4320310207","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33"},{"id":"https://openalex.org/F4320334164","display_name":"Simons Institute for the Theory of Computing, University of California Berkeley","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2990210896.pdf","grobid_xml":"https://content.openalex.org/works/W2990210896.grobid-xml"},"referenced_works_count":41,"referenced_works":["https://openalex.org/W21891419","https://openalex.org/W50486269","https://openalex.org/W107583932","https://openalex.org/W1505888104","https://openalex.org/W1542595278","https://openalex.org/W1553290137","https://openalex.org/W1850488217","https://openalex.org/W1938444229","https://openalex.org/W2020861455","https://openalex.org/W2054434031","https://openalex.org/W2103804774","https://openalex.org/W2115044435","https://openalex.org/W2119738618","https://openalex.org/W2129670787","https://openalex.org/W2130304665","https://openalex.org/W2488247662","https://openalex.org/W2489939061","https://openalex.org/W2545659366","https://openalex.org/W2604884452","https://openalex.org/W2724169821","https://openalex.org/W2769648743","https://openalex.org/W2787400134","https://openalex.org/W2788488350","https://openalex.org/W2794095813","https://openalex.org/W2809090039","https://openalex.org/W2890189120","https://openalex.org/W2901345551","https://openalex.org/W2907502549","https://openalex.org/W2912399346","https://openalex.org/W2918364912","https://openalex.org/W2942052807","https://openalex.org/W2943357710","https://openalex.org/W2944264312","https://openalex.org/W2962996309","https://openalex.org/W2963049774","https://openalex.org/W2963582321","https://openalex.org/W2963971282","https://openalex.org/W2965004202","https://openalex.org/W2970999177","https://openalex.org/W2992545808","https://openalex.org/W3198350258"],"related_works":["https://openalex.org/W4415796359","https://openalex.org/W4281712292","https://openalex.org/W3006052551","https://openalex.org/W2959895084","https://openalex.org/W3216847781","https://openalex.org/W2151661095","https://openalex.org/W4302011905","https://openalex.org/W1012918698","https://openalex.org/W3091123329","https://openalex.org/W4311432277"],"abstract_inverted_index":{"$Q$-learning":[0,93,106],"with":[1,36,79,94,109,210],"function":[2,19,81,96,111,143,146,179,212],"approximation":[3,20,82],"is":[4,40,183],"one":[5],"of":[6,17,121],"the":[7,15,30,66,80,129,145,177,197],"most":[8],"popular":[9],"methods":[10],"in":[11,29,77,144,203],"reinforcement":[12,207],"learning.":[13],"Though":[14],"idea":[16],"using":[18,117],"was":[21],"proposed":[22],"at":[23],"least":[24],"60":[25],"years":[26],"ago,":[27],"even":[28],"simplest":[31],"setup,":[32],"i.e,":[33],"approximating":[34],"$Q$-functions":[35],"linear":[37,95,110,178],"functions,":[38],"it":[39],"still":[41],"an":[42],"open":[43],"problem":[44],"on":[45,151,158],"how":[46,62,70],"to":[47,63,71,74,185],"design":[48],"a":[49,55,88,114,118,126,142,152,187],"provably":[50,89],"efficient":[51,90],"algorithm":[52,91,124],"that":[53,148],"learns":[54],"near-optimal":[56,115],"policy.":[57],"The":[58,84],"key":[59],"challenges":[60],"are":[61,166,200],"efficiently":[64],"explore":[65],"state":[67],"space":[68],"and":[69,164,205],"decide":[72],"when":[73],"stop":[75],"exploring":[76],"conjunction":[78],"scheme.":[83],"current":[85],"paper":[86],"presents":[87],"for":[92],"approximation.":[97,213],"Under":[98],"certain":[99],"regularity":[100],"assumptions,":[101],"our":[102,193],"algorithm,":[103],"Difference":[104],"Maximization":[105],"(DMQ),":[107],"combined":[108],"approximation,":[112],"returns":[113],"policy":[116],"polynomial":[119],"number":[120],"trajectories.":[122],"Our":[123],"introduces":[125],"new":[127],"notion,":[128],"Distribution":[130],"Shift":[131],"Error":[132],"Checking":[133],"(DSEC)":[134],"oracle.":[135],"This":[136],"oracle":[137,182],"tests":[138],"whether":[139],"there":[140],"exists":[141],"class":[147],"predicts":[149,156],"well":[150],"distribution":[153,160],"$\\mathcal{D}_1$,":[154],"but":[155],"poorly":[157],"another":[159],"$\\mathcal{D}_2$,":[161],"where":[162],"$\\mathcal{D}_1$":[163],"$\\mathcal{D}_2$":[165],"distributions":[167],"over":[168],"states":[169],"induced":[170],"by":[171],"two":[172],"different":[173],"exploration":[174],"policies.":[175],"For":[176],"class,":[180],"this":[181],"equivalent":[184],"solving":[186],"top":[188],"eigenvalue":[189],"problem.":[190],"We":[191],"believe":[192],"algorithmic":[194],"insights,":[195],"especially":[196],"DSEC":[198],"oracle,":[199],"also":[201],"useful":[202],"designing":[204],"analyzing":[206],"learning":[208],"algorithms":[209],"general":[211]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":7},{"year":2021,"cited_by_count":11},{"year":2020,"cited_by_count":22},{"year":2019,"cited_by_count":3}],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2019-12-05T00:00:00"}