{"id":"https://openalex.org/W3022833466","doi":"https://doi.org/10.24963/ijcai.2020/221","title":"Pitfalls of Learning a Reward Function Online","display_name":"Pitfalls of Learning a Reward Function Online","publication_year":2020,"publication_date":"2020-07-01","ids":{"openalex":"https://openalex.org/W3022833466","doi":"https://doi.org/10.24963/ijcai.2020/221","mag":"3022833466"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2020/221","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/221","pdf_url":"https://www.ijcai.org/proceedings/2020/0221.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.ijcai.org/proceedings/2020/0221.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5005897864","display_name":"Stuart Armstrong","orcid":"https://orcid.org/0000-0002-1385-3278"},"institutions":[{"id":"https://openalex.org/I283854653","display_name":"Machine Intelligence Research Institute","ror":"https://ror.org/01h4ass90","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I283854653"]},{"id":"https://openalex.org/I40120149","display_name":"University of Oxford","ror":"https://ror.org/052gg0110","country_code":"GB","type":"education","lineage":["https://openalex.org/I40120149"]}],"countries":["GB","US"],"is_corresponding":true,"raw_author_name":"Stuart Armstrong","raw_affiliation_strings":["Future of Humanity Institute, Oxford University, UK","Machine Intelligence Research Institute, Berkeley, USA","University of Oxford"],"affiliations":[{"raw_affiliation_string":"Future of Humanity Institute, Oxford University, UK","institution_ids":[]},{"raw_affiliation_string":"Machine Intelligence Research Institute, Berkeley, USA","institution_ids":["https://openalex.org/I283854653"]},{"raw_affiliation_string":"University of Oxford","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090592321","display_name":"Jan Leike","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090411","display_name":"DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Jan Leike","raw_affiliation_strings":["DeepMind, London, UK","Google,,,,,"],"affiliations":[{"raw_affiliation_string":"DeepMind, London, UK","institution_ids":["https://openalex.org/I4210090411"]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110938957","display_name":"Laurent Orseau","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I4210090411","display_name":"DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Laurent Orseau","raw_affiliation_strings":["DeepMind, London, UK","Google,,,,,"],"affiliations":[{"raw_affiliation_string":"DeepMind, London, UK","institution_ids":["https://openalex.org/I4210090411"]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5008987732","display_name":"Shane Legg","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I4210090411","display_name":"DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Shane Legg","raw_affiliation_strings":["DeepMind, London, UK","Google,,,,,"],"affiliations":[{"raw_affiliation_string":"DeepMind, London, UK","institution_ids":["https://openalex.org/I4210090411"]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5005897864"],"corresponding_institution_ids":["https://openalex.org/I283854653","https://openalex.org/I40120149"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.04668371,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1592","last_page":"1600"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9993000030517578,"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":0.9993000030517578,"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/T11182","display_name":"Auction Theory and Applications","score":0.9800000190734863,"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/T11031","display_name":"Game Theory and Applications","score":0.9732999801635742,"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/converse","display_name":"Converse","score":0.9126306772232056},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.6866245269775391},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6735968589782715},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.6676944494247437},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.6159531474113464},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5871062278747559},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.47441384196281433},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.10323327779769897}],"concepts":[{"id":"https://openalex.org/C2776809875","wikidata":"https://www.wikidata.org/wiki/Q1375963","display_name":"Converse","level":2,"score":0.9126306772232056},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.6866245269775391},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6735968589782715},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.6676944494247437},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.6159531474113464},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5871062278747559},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47441384196281433},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.10323327779769897},{"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/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.24963/ijcai.2020/221","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/221","pdf_url":"https://www.ijcai.org/proceedings/2020/0221.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2004.13654","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2004.13654","pdf_url":"https://arxiv.org/pdf/2004.13654","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"mag:3022833466","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2004.13654.pdf","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2004.13654","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2004.13654","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.24963/ijcai.2020/221","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2020/221","pdf_url":"https://www.ijcai.org/proceedings/2020/0221.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320317153","display_name":"DeepMind","ror":"https://ror.org/00971b260"},{"id":"https://openalex.org/F4320319993","display_name":"Leverhulme Trust","ror":"https://ror.org/012mzw131"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3022833466.pdf","grobid_xml":"https://content.openalex.org/works/W3022833466.grobid-xml"},"referenced_works_count":31,"referenced_works":["https://openalex.org/W41554520","https://openalex.org/W1555931943","https://openalex.org/W1583953806","https://openalex.org/W1594849649","https://openalex.org/W1669437150","https://openalex.org/W1999874108","https://openalex.org/W2061562262","https://openalex.org/W2098584016","https://openalex.org/W2257176436","https://openalex.org/W2312609093","https://openalex.org/W2383406194","https://openalex.org/W2410842990","https://openalex.org/W2573081168","https://openalex.org/W2575120333","https://openalex.org/W2626804490","https://openalex.org/W2752099845","https://openalex.org/W2768908787","https://openalex.org/W2770150859","https://openalex.org/W2889659926","https://openalex.org/W2951952563","https://openalex.org/W2955240493","https://openalex.org/W2962943921","https://openalex.org/W2963289505","https://openalex.org/W2963646405","https://openalex.org/W2964263543","https://openalex.org/W2968771440","https://openalex.org/W6634004297","https://openalex.org/W6650265722","https://openalex.org/W6679860557","https://openalex.org/W6680885310","https://openalex.org/W6812846395"],"related_works":["https://openalex.org/W3034558260","https://openalex.org/W3130606977","https://openalex.org/W1607218107","https://openalex.org/W2901707424","https://openalex.org/W3034769194","https://openalex.org/W1453801241","https://openalex.org/W2949969799","https://openalex.org/W1130790960","https://openalex.org/W1753708795","https://openalex.org/W3214285226","https://openalex.org/W3131041265","https://openalex.org/W2064306130","https://openalex.org/W3115631912","https://openalex.org/W2998135952","https://openalex.org/W2900406861","https://openalex.org/W2400719195","https://openalex.org/W2950172727","https://openalex.org/W2902143777","https://openalex.org/W1565527633","https://openalex.org/W2920768993"],"abstract_inverted_index":{"In":[0],"some":[1],"agent":[2,9,45,114],"designs":[3],"like":[4],"inverse":[5],"reinforcement":[6],"learning":[7,41,74,118,139,143],"an":[8,151],"needs":[10],"to":[11,80,86,130],"learn":[12],"its":[13],"own":[14],"reward":[15,19,49,99,125],"function.":[16],"Learning":[17],"the":[18,44,48,56,73,87,107,113,117,121,137,146,159,167],"function":[20,50,126],"and":[21,51,89,157],"optimising":[22],"for":[23,53],"it":[24,54],"are":[25,93],"typically":[26],"two":[27,104],"different":[28,33],"processes,":[29],"usually":[30],"performed":[31],"at":[32,55],"stages.":[34],"We":[35,59,101,148],"consider":[36],"a":[37,65,124],"continual":[38],"(``one":[39],"life'')":[40],"approach":[42],"where":[43],"both":[46],"learns":[47],"optimises":[52],"same":[57],"time.":[58],"show":[60,149],"that":[61,92,127,150],"this":[62],"comes":[63],"with":[64],"number":[66],"of":[67,123,161],"pitfalls,":[68],"such":[69],"as":[70],"deliberately":[71],"manipulating":[72],"process":[75,119,140,153],"in":[76,120],"one":[77],"direction,":[78],"refusing":[79],"learn,":[81],"``learning''":[82],"facts":[83,144],"already":[84],"known":[85],"agent,":[88],"making":[90],"decisions":[91],"strictly":[94],"dominated":[95],"(for":[96],"all":[97],"relevant":[98],"functions).":[100],"formally":[102],"introduce":[103],"desirable":[105],"properties:":[106],"first":[108],"is":[109,128,134,154,164,169],"`unriggability',":[110],"which":[111],"prevents":[112],"from":[115],"steering":[116],"direction":[122],"easier":[129],"optimise.":[131],"The":[132],"second":[133],"`uninfluenceability',":[135],"whereby":[136],"reward-function":[138],"operates":[141],"by":[142],"about":[145],"environment.":[147],"uninfluenceable":[152],"automatically":[155],"unriggable,":[156],"if":[158],"set":[160],"possible":[162],"environments":[163],"sufficiently":[165],"large,":[166],"converse":[168],"true":[170],"too.":[171]},"counts_by_year":[],"updated_date":"2026-03-10T16:38:18.471706","created_date":"2025-10-10T00:00:00"}