{"id":"https://openalex.org/W7124265117","doi":"https://doi.org/10.65109/xvvy9199","title":"Rethinking Out-of-Distribution Detection for Reinforcement Learning: Advancing Methods for Evaluation and Detection","display_name":"Rethinking Out-of-Distribution Detection for Reinforcement Learning: Advancing Methods for Evaluation and Detection","publication_year":2024,"publication_date":"2024-05-06","ids":{"openalex":"https://openalex.org/W7124265117","doi":"https://doi.org/10.65109/xvvy9199"},"language":"en","primary_location":{"id":"doi:10.65109/xvvy9199","doi":"https://doi.org/10.65109/xvvy9199","is_oa":false,"landing_page_url":"https://doi.org/10.65109/xvvy9199","pdf_url":null,"source":{"id":"https://openalex.org/S4306417597","display_name":"Adaptive Agents and Multi-Agents Systems","issn_l":"1548-8403","issn":["1548-8403"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Joint Conference on Autonomous Agents and Multiagent Systems","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2404.07099","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069707664","display_name":"Linas Nasvytis","orcid":"https://orcid.org/0009-0009-3910-2171","observed_orcids":["https://orcid.org/0009-0009-3910-2171"]},"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":"Linas Nasvytis","raw_affiliation_strings":["Harvard University, Cambridge, MA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University, Cambridge, MA, USA","institution_ids":["https://openalex.org/I136199984"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004390703","display_name":"Kai Jappe Sandbrink","orcid":"https://orcid.org/0000-0003-0360-075X","observed_orcids":["https://orcid.org/0000-0003-0360-075X"]},"institutions":[{"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"],"is_corresponding":false,"raw_author_name":"Kai Sandbrink","raw_affiliation_strings":["University of Oxford, Oxford, United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Oxford, Oxford, United Kingdom","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123081485","display_name":"Jakob Foerster","orcid":null,"observed_orcids":[]},"institutions":[{"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"],"is_corresponding":false,"raw_author_name":"Jakob Foerster","raw_affiliation_strings":["University of Oxford, Oxford, United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Oxford, Oxford, United Kingdom","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029454365","display_name":"Tim Franzmeyer","orcid":null,"observed_orcids":[]},"institutions":[{"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"],"is_corresponding":false,"raw_author_name":"Tim Franzmeyer","raw_affiliation_strings":["University of Oxford, Oxford, United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Oxford, Oxford, United Kingdom","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5038213642","display_name":"Christian A. Schroeder de Witt","orcid":null,"observed_orcids":[]},"institutions":[{"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"],"is_corresponding":false,"raw_author_name":"Christian Schroeder de Witt","raw_affiliation_strings":["University of Oxford, Oxford, United Kingdom"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Oxford, Oxford, United Kingdom","institution_ids":["https://openalex.org/I40120149"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.3679,"has_fulltext":true,"cited_by_count":16,"citation_normalized_percentile":{"value":0.91903411,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":88,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1445","last_page":"1453"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.5116999745368958,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.5116999745368958,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.40310001373291016,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.06459999829530716,"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/out-of-distribution-detection","display_name":"out-of-distribution detection","score":1.0},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"reinforcement learning","score":0.9639999866485596},{"id":"https://openalex.org/keywords/generalization-to-unseen-environments","display_name":"generalization to unseen environments","score":0.7670000195503235},{"id":"https://openalex.org/keywords/temporal-autocorrelation","display_name":"temporal autocorrelation","score":0.7900000214576721},{"id":"https://openalex.org/keywords/anomaly-detection","display_name":"anomaly detection","score":0.8320000171661377},{"id":"https://openalex.org/keywords/dexter","display_name":"Dexter","score":0.7889999747276306},{"id":"https://openalex.org/keywords/change-point-detection","display_name":"change point detection","score":0.859000027179718},{"id":"https://openalex.org/keywords/benchmark-scenarios","display_name":"benchmark scenarios","score":0.5979999899864197},{"id":"https://openalex.org/keywords/agent-environment-interaction","display_name":"agent-environment interaction","score":0.6919999718666077}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7109000086784363},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.7085999846458435},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6876999735832214},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6381999850273132},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.626800000667572},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6175000071525574},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.548799991607666},{"id":"https://openalex.org/C2780719617","wikidata":"https://www.wikidata.org/wiki/Q1030752","display_name":"Salient","level":2,"score":0.46389999985694885},{"id":"https://openalex.org/C94915269","wikidata":"https://www.wikidata.org/wiki/Q1834857","display_name":"Detector","level":2,"score":0.3944999873638153},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3864000141620636},{"id":"https://openalex.org/C2775941552","wikidata":"https://www.wikidata.org/wiki/Q25212305","display_name":"Isolation (microbiology)","level":2,"score":0.37400001287460327},{"id":"https://openalex.org/C151406439","wikidata":"https://www.wikidata.org/wiki/Q186588","display_name":"Time series","level":2,"score":0.37139999866485596},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.337799996137619},{"id":"https://openalex.org/C143724316","wikidata":"https://www.wikidata.org/wiki/Q312468","display_name":"Series (stratigraphy)","level":2,"score":0.3345000147819519},{"id":"https://openalex.org/C547195049","wikidata":"https://www.wikidata.org/wiki/Q1725664","display_name":"Terminology","level":2,"score":0.32269999384880066},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.32260000705718994},{"id":"https://openalex.org/C64869954","wikidata":"https://www.wikidata.org/wiki/Q1859747","display_name":"False positive paradox","level":2,"score":0.321399986743927},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.3199000060558319},{"id":"https://openalex.org/C2778473407","wikidata":"https://www.wikidata.org/wiki/Q1459574","display_name":"Compendium","level":2,"score":0.3142000138759613},{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.2985000014305115},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.29829999804496765},{"id":"https://openalex.org/C45942800","wikidata":"https://www.wikidata.org/wiki/Q245652","display_name":"Ensemble learning","level":2,"score":0.2766999900341034},{"id":"https://openalex.org/C5297727","wikidata":"https://www.wikidata.org/wiki/Q786970","display_name":"Autocorrelation","level":2,"score":0.2621000111103058}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.65109/xvvy9199","doi":"https://doi.org/10.65109/xvvy9199","is_oa":false,"landing_page_url":"https://doi.org/10.65109/xvvy9199","pdf_url":null,"source":{"id":"https://openalex.org/S4306417597","display_name":"Adaptive Agents and Multi-Agents Systems","issn_l":"1548-8403","issn":["1548-8403"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Joint Conference on Autonomous Agents and Multiagent Systems","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2404.07099","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2404.07099","pdf_url":"https://arxiv.org/pdf/2404.07099","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2404.07099","doi":"https://doi.org/10.48550/arxiv.2404.07099","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2404.07099","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:2404.07099","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2404.07099","pdf_url":"https://arxiv.org/pdf/2404.07099","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":"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/F4320317463","display_name":"Future of Life Institute","ror":"https://ror.org/03ey0bd98"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7124265117.pdf","grobid_xml":"https://content.openalex.org/works/W7124265117.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W2051903196","https://openalex.org/W2736601468","https://openalex.org/W2898436992","https://openalex.org/W2945353373","https://openalex.org/W2994722919","https://openalex.org/W2997118265","https://openalex.org/W3037207827","https://openalex.org/W3178385351","https://openalex.org/W4200633036","https://openalex.org/W4286750943","https://openalex.org/W4386285856","https://openalex.org/W4399929809"],"related_works":["https://openalex.org/W7168362067","https://openalex.org/W4200633036","https://openalex.org/W7124320695","https://openalex.org/W4417328270","https://openalex.org/W3178385351","https://openalex.org/W4409365247","https://openalex.org/W4398186391","https://openalex.org/W3210963246","https://openalex.org/W7165029028","https://openalex.org/W4402352426"],"abstract_inverted_index":{"While":[0],"reinforcement":[1],"learning":[2,76],"(RL)":[3],"algorithms":[4,182],"have":[5,48,105],"been":[6,106],"successfully":[7],"applied":[8],"across":[9,194],"numerous":[10],"sequential":[11],"decision-making":[12],"problems,":[13],"their":[14,52,113],"generalization":[15],"to":[16,115,133,183,201],"unforeseen":[17],"testing":[18],"environments":[19],"remains":[20],"a":[21,58,143],"significant":[22],"concern.":[23],"In":[24],"this":[25,139],"paper,":[26],"we":[27,141,150],"study":[28],"the":[29,71,97,109],"problem":[30],"of":[31,60,96,156,179],"out-of-distribution":[32],"(OOD)":[33],"detection":[34,64],"in":[35,51,65,108],"RL,":[36,66],"which":[37,67,87,149],"focuses":[38],"on":[39],"identifying":[40],"situations":[41],"at":[42],"test":[43],"time":[44,165,171],"that":[45,102,126,188],"RL":[46],"agents":[47],"not":[49,131],"encountered":[50],"training":[53],"environments.":[54],"We":[55,78,100,186],"first":[56],"propose":[57,142],"clarification":[59],"terminology":[61],"for":[62,84,146],"OOD":[63,85,128,147,204],"aligns":[68],"it":[69],"with":[70,90],"literature":[72],"from":[73,211],"other":[74],"machine":[75],"domains.":[77],"then":[79,175],"present":[80],"new":[81],"benchmark":[82,195],"scenarios":[83,104],"detection,":[86,148],"introduce":[88],"anomalies":[89,193],"temporal":[91],"autocorrelation":[92],"into":[93],"different":[94],"components":[95],"agent-environment":[98],"loop.":[99],"argue":[101],"such":[103,135],"understudied":[107],"current":[110],"literature,":[111],"despite":[112],"relevance":[114],"real-world":[116],"situations.":[117],"Confirming":[118],"our":[119,122],"theoretical":[120],"predictions,":[121],"experimental":[123],"results":[124],"suggest":[125],"state-of-the-art":[127,203],"detectors":[129,205,209],"are":[130],"able":[132],"identify":[134,192],"anomalies.":[136,185],"To":[137],"address":[138],"problem,":[140],"novel":[144],"method":[145],"call":[151],"DEXTER":[152,168,189],"(Detection":[153],"via":[154],"Extraction":[155],"Time":[157],"Series":[158],"Representations).":[159],"By":[160],"treating":[161],"environment":[162],"observations":[163],"as":[164],"series":[166,172],"data,":[167],"extracts":[169],"salient":[170],"features,":[173],"and":[174,206],"leverages":[176],"an":[177],"ensemble":[178],"isolation":[180],"forest":[181],"detect":[184],"find":[187],"can":[190],"reliably":[191],"scenarios,":[196],"exhibiting":[197],"superior":[198],"performance":[199],"compared":[200],"both":[202],"high-dimensional":[207],"changepoint":[208],"adopted":[210],"statistics.":[212]},"counts_by_year":[{"year":2026,"cited_by_count":7},{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":1}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-01-16T00:00:00"}