{"id":"https://openalex.org/W2075268401","doi":"https://doi.org/10.1145/1553374.1553501","title":"Fast gradient-descent methods for temporal-difference learning with linear function approximation","display_name":"Fast gradient-descent methods for temporal-difference learning with linear function approximation","publication_year":2009,"publication_date":"2009-06-14","ids":{"openalex":"https://openalex.org/W2075268401","doi":"https://doi.org/10.1145/1553374.1553501","mag":"2075268401"},"language":"en","primary_location":{"id":"doi:10.1145/1553374.1553501","is_oa":false,"landing_page_url":"https://doi.org/10.1145/1553374.1553501","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 26th Annual International Conference on Machine Learning","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5004923102","display_name":"Richard S. Sutton","orcid":"https://orcid.org/0000-0002-3679-3415"},"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"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Richard S. Sutton","raw_affiliation_strings":["University of Alberta, Edmonton, Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Alberta, Edmonton, Canada","institution_ids":["https://openalex.org/I154425047"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109219313","display_name":"Hamid Reza Maei","orcid":null},"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"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Hamid Reza Maei","raw_affiliation_strings":["University of Alberta, Edmonton, Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Alberta, Edmonton, Canada","institution_ids":["https://openalex.org/I154425047"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065836447","display_name":"Doina Precup","orcid":null},"institutions":[{"id":"https://openalex.org/I5023651","display_name":"McGill University","ror":"https://ror.org/01pxwe438","country_code":"CA","type":"education","lineage":["https://openalex.org/I5023651"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Doina Precup","raw_affiliation_strings":["McGill University, Montreal, Canada","McGill University; Montreal; Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"McGill University, Montreal, Canada","institution_ids":["https://openalex.org/I5023651"]},{"raw_affiliation_string":"McGill University; Montreal; Canada","institution_ids":["https://openalex.org/I5023651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038163398","display_name":"Shalabh Bhatnagar","orcid":"https://orcid.org/0000-0001-7644-3914"},"institutions":[{"id":"https://openalex.org/I59270414","display_name":"Indian Institute of Science Bangalore","ror":"https://ror.org/04dese585","country_code":"IN","type":"education","lineage":["https://openalex.org/I59270414"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Shalabh Bhatnagar","raw_affiliation_strings":["Indian Institute of Science, Bangalore, India"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Indian Institute of Science, Bangalore, India","institution_ids":["https://openalex.org/I59270414"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091771290","display_name":"David Silver","orcid":"https://orcid.org/0000-0002-5197-2892"},"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"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"David Silver","raw_affiliation_strings":["University of Alberta, Edmonton, Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Alberta, Edmonton, Canada","institution_ids":["https://openalex.org/I154425047"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069856068","display_name":"Csaba Szepesv\u00e1ri","orcid":"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"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Csaba Szepesv\u00e1ri","raw_affiliation_strings":["University of Alberta, Edmonton, Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Alberta, Edmonton, Canada","institution_ids":["https://openalex.org/I154425047"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5085248782","display_name":"Eric Wiewiora","orcid":null},"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"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Eric Wiewiora","raw_affiliation_strings":["University of Alberta, Edmonton, Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Alberta, Edmonton, Canada","institution_ids":["https://openalex.org/I154425047"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":19.0032,"has_fulltext":false,"cited_by_count":536,"citation_normalized_percentile":{"value":0.99676847,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"993","last_page":"1000"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9994999766349792,"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":0.9994999766349792,"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/T12676","display_name":"Machine Learning and ELM","score":0.9980000257492065,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9976000189781189,"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/gradient-descent","display_name":"Gradient descent","score":0.6202964186668396},{"id":"https://openalex.org/keywords/temporal-difference-learning","display_name":"Temporal difference learning","score":0.5636188387870789},{"id":"https://openalex.org/keywords/linear-approximation","display_name":"Linear approximation","score":0.5612518787384033},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"Stochastic gradient descent","score":0.5294746160507202},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5087109208106995},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.4744761288166046},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.456521213054657},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.34878551959991455},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.33846864104270935},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.24007448554039001},{"id":"https://openalex.org/keywords/nonlinear-system","display_name":"Nonlinear system","score":0.14817219972610474},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.1455630362033844},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.13494369387626648},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.08628171682357788}],"concepts":[{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.6202964186668396},{"id":"https://openalex.org/C196340769","wikidata":"https://www.wikidata.org/wiki/Q7698910","display_name":"Temporal difference learning","level":3,"score":0.5636188387870789},{"id":"https://openalex.org/C160824197","wikidata":"https://www.wikidata.org/wiki/Q2071054","display_name":"Linear approximation","level":3,"score":0.5612518787384033},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5294746160507202},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5087109208106995},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.4744761288166046},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.456521213054657},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.34878551959991455},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.33846864104270935},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.24007448554039001},{"id":"https://openalex.org/C158622935","wikidata":"https://www.wikidata.org/wiki/Q660848","display_name":"Nonlinear system","level":2,"score":0.14817219972610474},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.1455630362033844},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.13494369387626648},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.08628171682357788},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"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/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":6,"locations":[{"id":"doi:10.1145/1553374.1553501","is_oa":false,"landing_page_url":"https://doi.org/10.1145/1553374.1553501","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 26th Annual International Conference on Machine Learning","raw_type":"proceedings-article"},{"id":"pmh:oai:CiteSeerX.psu:10.1.1.149.5674","is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.149.5674","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://www.cs.mcgill.ca/~icml2009/papers/546.pdf","raw_type":"text"},{"id":"pmh:oai:CiteSeerX.psu:10.1.1.150.1759","is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.150.1759","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://www.sztaki.hu/~szcsaba/papers/GTD-ICML09.pdf","raw_type":"text"},{"id":"pmh:oai:CiteSeerX.psu:10.1.1.160.6170","is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.160.6170","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://www.cs.ualberta.ca/~sutton/papers/SMPBSSW-09.pdf","raw_type":"text"},{"id":"pmh:oai:CiteSeerX.psu:10.1.1.161.1442","is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.161.1442","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://www.cs.ualberta.ca/~silver/research/publications/files/SMPBSSW-09.pdf","raw_type":"text"},{"id":"pmh:oai:eprints.ucl.ac.uk.OAI2:1368112","is_oa":false,"landing_page_url":"http://discovery.ucl.ac.uk/1368112/","pdf_url":null,"source":{"id":"https://openalex.org/S4306400024","display_name":"UCL Discovery (University College London)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I45129253","host_organization_name":"University College London","host_organization_lineage":["https://openalex.org/I45129253"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"In: (2009)","raw_type":"Proceedings paper"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.5099999904632568,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W86816279","https://openalex.org/W359568995","https://openalex.org/W1583837637","https://openalex.org/W1597317312","https://openalex.org/W1600046456","https://openalex.org/W1646707810","https://openalex.org/W1778554682","https://openalex.org/W2065134213","https://openalex.org/W2068778050","https://openalex.org/W2071983464","https://openalex.org/W2072931156","https://openalex.org/W2094364653","https://openalex.org/W2100677568","https://openalex.org/W2104753538","https://openalex.org/W2109910161","https://openalex.org/W2115253045","https://openalex.org/W2132351269","https://openalex.org/W2139418546","https://openalex.org/W2150459019","https://openalex.org/W2158091072","https://openalex.org/W6638026421"],"related_works":["https://openalex.org/W2895097035","https://openalex.org/W4206903459","https://openalex.org/W2754816816","https://openalex.org/W4366280654","https://openalex.org/W3160167280","https://openalex.org/W3020853991","https://openalex.org/W4362706668","https://openalex.org/W4231621013","https://openalex.org/W3171021120","https://openalex.org/W3008318776"],"abstract_inverted_index":{"Sutton,":[0],"Szepesv\u00e1ri":[1],"and":[2,18,21,85,98,145],"Maei":[3],"(2009)":[4],"recently":[5],"introduced":[6],"the":[7,28,31,122,155],"first":[8,80],"temporal-difference":[9],"learning":[10,156,177],"algorithm":[11,40,160,169],"compatible":[12],"with":[13,75,116,151,178],"both":[14],"linear":[15,51,114,173],"function":[16,32,97],"approximation":[17],"off-policy":[19,176],"training,":[20],"whose":[22],"complexity":[23],"scales":[24],"only":[25,184],"linearly":[26],"in":[27,146,181],"size":[29],"of":[30,158,165],"approximator.":[33],"Although":[34],"their":[35],"gradient":[36,117],"temporal":[37],"difference":[38],"(GTD)":[39],"converges":[41,99],"reliably,":[42],"it":[43],"can":[44],"be":[45],"very":[46],"slow":[47],"compared":[48],"to":[49,163,171,175],"conventional":[50,108,127,166],"TD":[52,57,115,128,174],"(on":[53],"on-policy":[54],"problems":[55,144],"where":[56],"is":[58,83,135],"convergent),":[59],"calling":[60],"into":[61],"question":[62],"its":[63],"practical":[64],"utility.":[65],"In":[66,138],"this":[67,159],"paper":[68],"we":[69],"introduce":[70],"two":[71],"new":[72,112],"related":[73],"algorithms":[74],"better":[76],"convergence":[77],"rates.":[78],"The":[79,110],"algorithm,":[81,113],"GTD2,":[82],"derived":[84],"proved":[86],"convergent":[87],"just":[88],"as":[89,105,107,126],"GTD":[90],"was,":[91],"but":[92],"uses":[93,121],"a":[94,147,152],"different":[95],"objective":[96],"significantly":[100],"faster":[101],"(but":[102],"still":[103],"not":[104],"fast":[106],"TD).":[109],"second":[111],"correction,":[118],"or":[119],"TDC,":[120],"same":[123],"update":[124],"rule":[125],"except":[129],"for":[130],"an":[131],"additional":[132],"term":[133],"which":[134],"initially":[136],"zero.":[137],"our":[139],"experiments":[140],"on":[141],"small":[142],"test":[143],"Computer":[148],"Go":[149],"application":[150],"million":[153],"features,":[154],"rate":[157],"was":[161],"comparable":[162],"that":[164],"TD.":[167],"This":[168],"appears":[170],"extend":[172],"no":[179],"penalty":[180],"performance":[182],"while":[183],"doubling":[185],"computational":[186],"requirements.":[187]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":16},{"year":2024,"cited_by_count":23},{"year":2023,"cited_by_count":21},{"year":2022,"cited_by_count":32},{"year":2021,"cited_by_count":64},{"year":2020,"cited_by_count":59},{"year":2019,"cited_by_count":63},{"year":2018,"cited_by_count":36},{"year":2017,"cited_by_count":33},{"year":2016,"cited_by_count":24},{"year":2015,"cited_by_count":24},{"year":2014,"cited_by_count":32},{"year":2013,"cited_by_count":26},{"year":2012,"cited_by_count":33}],"updated_date":"2026-08-01T09:00:35.917206","created_date":"2025-10-10T00:00:00"}