{"id":"https://openalex.org/W3131973708","doi":"https://doi.org/10.48550/arxiv.2102.12611","title":"Improved Regret Bound and Experience Replay in Regularized Policy Iteration","display_name":"Improved Regret Bound and Experience Replay in Regularized Policy Iteration","publication_year":2021,"publication_date":"2021-02-25","ids":{"openalex":"https://openalex.org/W3131973708","doi":"https://doi.org/10.48550/arxiv.2102.12611","mag":"3131973708"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2102.12611","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2102.12611","pdf_url":"https://arxiv.org/pdf/2102.12611","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/2102.12611","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5043371034","display_name":"Nevena Lazic","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I154425047","display_name":"University of Alberta","ror":"https://ror.org/0160cpw27","country_code":"CA","type":"education","lineage":["https://openalex.org/I154425047"]},{"id":"https://openalex.org/I4210090411","display_name":"Google DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]}],"countries":["CA","GB"],"is_corresponding":false,"raw_author_name":"Lazic, Nevena","raw_affiliation_strings":["University of Alberta","DeepMind"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Alberta","institution_ids":["https://openalex.org/I154425047"]},{"raw_affiliation_string":"DeepMind","institution_ids":["https://openalex.org/I4210090411"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001139821","display_name":"Dong Yin","orcid":"https://orcid.org/0000-0002-2358-0816","observed_orcids":["https://orcid.org/0000-0002-2358-0816"]},"institutions":[{"id":"https://openalex.org/I154425047","display_name":"University of Alberta","ror":"https://ror.org/0160cpw27","country_code":"CA","type":"education","lineage":["https://openalex.org/I154425047"]},{"id":"https://openalex.org/I4210090411","display_name":"Google DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]}],"countries":["CA","GB"],"is_corresponding":false,"raw_author_name":"Yin, Dong","raw_affiliation_strings":["University of Alberta","DeepMind"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Alberta","institution_ids":["https://openalex.org/I154425047"]},{"raw_affiliation_string":"DeepMind","institution_ids":["https://openalex.org/I4210090411"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047010941","display_name":"Yasin Abbasi-Yadkori","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I1306409833","display_name":"Adobe Systems (United States)","ror":"https://ror.org/059tvcg64","country_code":"US","type":"company","lineage":["https://openalex.org/I1306409833"]},{"id":"https://openalex.org/I154425047","display_name":"University of Alberta","ror":"https://ror.org/0160cpw27","country_code":"CA","type":"education","lineage":["https://openalex.org/I154425047"]}],"countries":["CA","US"],"is_corresponding":false,"raw_author_name":"Abbasi-Yadkori, Yasin","raw_affiliation_strings":["University of Alberta","Adobe Research#TAB#"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Alberta","institution_ids":["https://openalex.org/I154425047"]},{"raw_affiliation_string":"Adobe Research#TAB#","institution_ids":["https://openalex.org/I1306409833"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5069856068","display_name":"Csaba Szepesv\u00e1ri","orcid":"https://orcid.org/0000-0002-9286-2892","observed_orcids":["https://orcid.org/0000-0002-9286-2892"]},"institutions":[{"id":"https://openalex.org/I154425047","display_name":"University of Alberta","ror":"https://ror.org/0160cpw27","country_code":"CA","type":"education","lineage":["https://openalex.org/I154425047"]},{"id":"https://openalex.org/I4210090411","display_name":"Google DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]}],"countries":["CA","GB"],"is_corresponding":false,"raw_author_name":"Szepesvari, Csaba","raw_affiliation_strings":["University of Alberta","DeepMind/University of Alberta"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Alberta","institution_ids":["https://openalex.org/I154425047"]},{"raw_affiliation_string":"DeepMind/University of Alberta","institution_ids":["https://openalex.org/I154425047","https://openalex.org/I4210090411"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"6032","last_page":"6042"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9961000084877014,"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":0.9961000084877014,"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":0.0031999999191612005,"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":0.0003000000142492354,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/regret-bounds","display_name":"regret bounds","score":0.9399999976158142},{"id":"https://openalex.org/keywords/function-approximation","display_name":"function approximation","score":0.9380000233650208},{"id":"https://openalex.org/keywords/experience-replay","display_name":"experience replay","score":0.7390000224113464},{"id":"https://openalex.org/keywords/q-function","display_name":"Q-function","score":0.6499999761581421},{"id":"https://openalex.org/keywords/linear-function-approximation","display_name":"linear function approximation","score":0.8429999947547913},{"id":"https://openalex.org/keywords/neural-network-function-approximation","display_name":"neural network function approximation","score":0.8769999742507935},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"reinforcement learning","score":0.5910000205039978}],"concepts":[{"id":"https://openalex.org/C50817715","wikidata":"https://www.wikidata.org/wiki/Q79895177","display_name":"Regret","level":2,"score":0.9281700849533081},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.7746206521987915},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6050372123718262},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.6017375588417053},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5637738704681396},{"id":"https://openalex.org/C14646407","wikidata":"https://www.wikidata.org/wiki/Q1430750","display_name":"Bellman equation","level":2,"score":0.48874402046203613},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4489190876483917},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.44670426845550537},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.439574658870697},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.35369086265563965},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3216872215270996},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.317457914352417},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3150646984577179},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.12042436003684998},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"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/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:2102.12611","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2102.12611","pdf_url":"https://arxiv.org/pdf/2102.12611","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.2102.12611","doi":"https://doi.org/10.48550/arxiv.2102.12611","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2102.12611","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:3168197022","doi":null,"is_oa":false,"landing_page_url":"http://proceedings.mlr.press/v139/lazic21a/lazic21a.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306419644","display_name":"International Conference on Machine Learning","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"International Conference on Machine Learning","raw_type":null}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2102.12611","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2102.12611","pdf_url":"https://arxiv.org/pdf/2102.12611","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":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3131973708.pdf","grobid_xml":"https://content.openalex.org/works/W3131973708.grobid-xml"},"referenced_works_count":53,"referenced_works":["https://openalex.org/W1497026160","https://openalex.org/W1545748428","https://openalex.org/W1570963478","https://openalex.org/W1575592356","https://openalex.org/W1583833196","https://openalex.org/W1662803991","https://openalex.org/W1757796397","https://openalex.org/W1771410628","https://openalex.org/W1850488217","https://openalex.org/W1906772730","https://openalex.org/W1983161979","https://openalex.org/W2000850397","https://openalex.org/W2074680702","https://openalex.org/W2075672181","https://openalex.org/W2106261932","https://openalex.org/W2107411554","https://openalex.org/W2141559645","https://openalex.org/W2145339207","https://openalex.org/W2173248099","https://openalex.org/W2201581102","https://openalex.org/W2326968954","https://openalex.org/W2333017703","https://openalex.org/W2426267443","https://openalex.org/W2560647685","https://openalex.org/W2612467436","https://openalex.org/W2736601468","https://openalex.org/W2737492962","https://openalex.org/W2781585732","https://openalex.org/W2785738552","https://openalex.org/W2786907751","https://openalex.org/W2790409593","https://openalex.org/W2902625698","https://openalex.org/W2914234208","https://openalex.org/W2945496654","https://openalex.org/W2949608212","https://openalex.org/W2962724315","https://openalex.org/W2962803570","https://openalex.org/W2964000194","https://openalex.org/W2970720882","https://openalex.org/W2975395546","https://openalex.org/W2980753060","https://openalex.org/W2981073936","https://openalex.org/W2988787144","https://openalex.org/W3004970331","https://openalex.org/W3014137283","https://openalex.org/W3026615607","https://openalex.org/W3032949543","https://openalex.org/W3035397247","https://openalex.org/W3036362369","https://openalex.org/W3045472490","https://openalex.org/W3094105627","https://openalex.org/W3158744994","https://openalex.org/W4319988532"],"related_works":["https://openalex.org/W2158738729","https://openalex.org/W3045472490","https://openalex.org/W3158744994","https://openalex.org/W3020075106","https://openalex.org/W2276878381","https://openalex.org/W4393967375","https://openalex.org/W4318719340","https://openalex.org/W3004970331","https://openalex.org/W3020125231","https://openalex.org/W7100823339"],"abstract_inverted_index":{"In":[0],"this":[1,69,130],"work,":[2],"we":[3,94,108,115],"study":[4],"algorithms":[5],"for":[6,63,154],"learning":[7],"in":[8,68,83,141],"infinite-horizon":[9],"undiscounted":[10],"Markov":[11],"decision":[12],"processes":[13],"(MDPs)":[14],"with":[15,50,76,124],"function":[16,52,79],"approximation.":[17,53],"We":[18,127],"first":[19,58],"show":[20,128],"that":[21,93,129],"the":[22,26,48,57,98,101],"regret":[23,61],"analysis":[24,91],"of":[25,31,74,85,100,104,143],"Politex":[27,75],"algorithm":[28,67],"(a":[29],"version":[30],"regularized":[32],"policy":[33,159],"iteration)":[34],"can":[35],"be":[36],"sharpened":[37],"from":[38],"$O(T^{3/4})$":[39],"to":[40,96,133],"$O(\\sqrt{T})$":[41,60],"under":[42],"nearly":[43],"identical":[44],"assumptions,":[45],"and":[46,87],"instantiate":[47],"bound":[49,62],"linear":[51],"Our":[54,146],"result":[55],"provides":[56,149],"high-probability":[59],"a":[64,110,117,121,150],"computationally":[65],"efficient":[66,112],"setting.":[70],"The":[71],"exact":[72],"implementation":[73,113,138],"neural":[77],"network":[78],"approximation":[80],"is":[81],"inefficient":[82],"terms":[84,142],"memory":[86],"computation.":[88],"Since":[89],"our":[90],"suggests":[92],"need":[95],"approximate":[97],"average":[99],"action-value":[102],"functions":[103],"past":[105,125],"policies":[106],"well,":[107],"propose":[109],"simple":[111],"where":[114],"train":[116],"single":[118],"Q-function":[119],"on":[120],"replay":[122,157],"buffer":[123],"data.":[126],"often":[131],"leads":[132],"superior":[134],"performance":[135],"over":[136],"other":[137],"choices,":[139],"especially":[140],"wall-clock":[144],"time.":[145],"work":[147],"also":[148],"novel":[151],"theoretical":[152],"justification":[153],"using":[155],"experience":[156],"within":[158],"iteration":[160],"algorithms.":[161]},"counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":2}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2021-03-01T00:00:00"}