{"id":"https://openalex.org/W4398186391","doi":"https://doi.org/10.1145/3605098.3635931","title":"Can you trust your Agent? The Effect of Out-of-Distribution Detection on the Safety of Reinforcement Learning Systems","display_name":"Can you trust your Agent? The Effect of Out-of-Distribution Detection on the Safety of Reinforcement Learning Systems","publication_year":2024,"publication_date":"2024-04-08","ids":{"openalex":"https://openalex.org/W4398186391","doi":"https://doi.org/10.1145/3605098.3635931"},"language":"en","primary_location":{"id":"doi:10.1145/3605098.3635931","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3605098.3635931","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3605098.3635931","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 39th ACM/SIGAPP Symposium on Applied Computing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3605098.3635931","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5087368349","display_name":"Tom Haider","orcid":"https://orcid.org/0000-0001-6786-0361"},"institutions":[{"id":"https://openalex.org/I4210127444","display_name":"Fraunhofer Institute for Cognitive Systems","ror":"https://ror.org/02fez3815","country_code":"DE","type":"facility","lineage":["https://openalex.org/I4210127444","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Tom Haider","raw_affiliation_strings":["Fraunhofer IKS, Munich, Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer IKS, Munich, Germany","institution_ids":["https://openalex.org/I4210127444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050789675","display_name":"Karsten Roscher","orcid":"https://orcid.org/0000-0002-9458-104X"},"institutions":[{"id":"https://openalex.org/I4210127444","display_name":"Fraunhofer Institute for Cognitive Systems","ror":"https://ror.org/02fez3815","country_code":"DE","type":"facility","lineage":["https://openalex.org/I4210127444","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Karsten Roscher","raw_affiliation_strings":["Fraunhofer IKS, Munich, Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer IKS, Munich, Germany","institution_ids":["https://openalex.org/I4210127444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003599831","display_name":"Benjamin Herd","orcid":"https://orcid.org/0000-0001-6439-8845"},"institutions":[{"id":"https://openalex.org/I4210127444","display_name":"Fraunhofer Institute for Cognitive Systems","ror":"https://ror.org/02fez3815","country_code":"DE","type":"facility","lineage":["https://openalex.org/I4210127444","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Benjamin Herd","raw_affiliation_strings":["Fraunhofer IKS, Munich, Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer IKS, Munich, Germany","institution_ids":["https://openalex.org/I4210127444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068613985","display_name":"Felippe Schmoeller Roza","orcid":"https://orcid.org/0000-0001-6900-6601"},"institutions":[{"id":"https://openalex.org/I4210127444","display_name":"Fraunhofer Institute for Cognitive Systems","ror":"https://ror.org/02fez3815","country_code":"DE","type":"facility","lineage":["https://openalex.org/I4210127444","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Felippe Schmoeller Roza","raw_affiliation_strings":["Fraunhofer IKS, Munich, Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer IKS, Munich, Germany","institution_ids":["https://openalex.org/I4210127444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051274373","display_name":"Simon Burton","orcid":"https://orcid.org/0000-0001-9040-8752"},"institutions":[{"id":"https://openalex.org/I4210127444","display_name":"Fraunhofer Institute for Cognitive Systems","ror":"https://ror.org/02fez3815","country_code":"DE","type":"facility","lineage":["https://openalex.org/I4210127444","https://openalex.org/I4923324"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Simon Burton","raw_affiliation_strings":["Fraunhofer IKS, Munich, Germany"],"affiliations":[{"raw_affiliation_string":"Fraunhofer IKS, Munich, Germany","institution_ids":["https://openalex.org/I4210127444"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5087368349"],"corresponding_institution_ids":["https://openalex.org/I4210127444"],"apc_list":null,"apc_paid":null,"fwci":1.9219,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.86914144,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1569","last_page":"1578"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12423","display_name":"Software Reliability and Analysis Research","score":0.9897000193595886,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/T12423","display_name":"Software Reliability and Analysis Research","score":0.9897000193595886,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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":0.9771999716758728,"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/T10917","display_name":"Smart Grid Security and Resilience","score":0.9732000231742859,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.753122091293335},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.601119875907898},{"id":"https://openalex.org/keywords/reinforcement","display_name":"Reinforcement","score":0.49772313237190247},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.34827423095703125},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.33464041352272034},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2775661051273346},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.11172521114349365},{"id":"https://openalex.org/keywords/social-psychology","display_name":"Social psychology","score":0.06807553768157959}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.753122091293335},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.601119875907898},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.49772313237190247},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.34827423095703125},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.33464041352272034},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2775661051273346},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.11172521114349365},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.06807553768157959}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3605098.3635931","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3605098.3635931","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3605098.3635931","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 39th ACM/SIGAPP Symposium on Applied Computing","raw_type":"proceedings-article"},{"id":"pmh:oai:eprints.whiterose.ac.uk:230127","is_oa":true,"landing_page_url":null,"pdf_url":"https://eprints.whiterose.ac.uk/id/eprint/230127/1/3605098.3635931.pdf","source":{"id":"https://openalex.org/S4306400854","display_name":"White Rose Research Online (University of Leeds, The University of Sheffield, University of York)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I2800616092","host_organization_name":"White Rose University Consortium","host_organization_lineage":["https://openalex.org/I2800616092"],"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":"NonPeerReviewed"},{"id":"pmh:oai:publica.fraunhofer.de:publica/469635","is_oa":true,"landing_page_url":"https://publica.fraunhofer.de/handle/publica/469635","pdf_url":null,"source":{"id":"https://openalex.org/S4306400318","display_name":"Fraunhofer-Publica (Fraunhofer-Gesellschaft)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4923324","host_organization_name":"Fraunhofer-Gesellschaft","host_organization_lineage":["https://openalex.org/I4923324"],"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":"conference paper"}],"best_oa_location":{"id":"doi:10.1145/3605098.3635931","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3605098.3635931","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3605098.3635931","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 39th ACM/SIGAPP Symposium on Applied Computing","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.800000011920929}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4398186391.pdf"},"referenced_works_count":14,"referenced_works":["https://openalex.org/W1579747287","https://openalex.org/W1723619723","https://openalex.org/W2122646361","https://openalex.org/W2750631532","https://openalex.org/W2886168462","https://openalex.org/W2963575966","https://openalex.org/W2966735560","https://openalex.org/W2985970291","https://openalex.org/W3003931103","https://openalex.org/W3097884524","https://openalex.org/W3110281108","https://openalex.org/W3195968524","https://openalex.org/W4234266781","https://openalex.org/W4362671719"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2920061524","https://openalex.org/W4310083477","https://openalex.org/W2328553770","https://openalex.org/W1977959518","https://openalex.org/W2038908348","https://openalex.org/W2107890255","https://openalex.org/W2106552856","https://openalex.org/W2145821588"],"abstract_inverted_index":{"Deep":[0],"Reinforcement":[1],"Learning":[2,168],"(RL)":[3],"has":[4,18],"the":[5,9,40,46,62,81,100,107,110,129,136],"potential":[6,152],"to":[7,22,29,128,186,203],"revolutionize":[8],"automation":[10],"of":[11,26,39,48,109,131,166,173,210],"complex":[12],"sequential":[13],"decision-making":[14],"problems.":[15],"Although":[16],"it":[17,184],"been":[19],"successfully":[20],"applied":[21],"a":[23,155,171],"wide":[24],"range":[25],"tasks,":[27],"deployment":[28],"real-world":[30],"settings":[31],"remains":[32],"challenging":[33],"and":[34,150,180,196],"is":[35,45,114,185,212],"often":[36],"limited.":[37],"One":[38],"main":[41],"reasons":[42],"for":[43,51,147,154,163],"this":[44,120],"lack":[47],"safety":[49,108,130,157,164],"guarantees":[50],"conventional":[52],"RL":[53,98,132],"algorithms,":[54],"especially":[55],"in":[56,99,194],"situations":[57],"that":[58],"substantially":[59],"differ":[60],"from":[61,189],"learning":[63],"environment.":[64],"In":[65,119,170],"such":[66],"situations,":[67],"state-of-the-art":[68],"systems":[69,133],"will":[70],"fail":[71],"silently,":[72],"producing":[73],"action":[74],"sequences":[75],"without":[76],"signalizing":[77],"any":[78],"uncertainty":[79],"regarding":[80],"current":[82],"input.":[83],"Recent":[84],"works":[85],"have":[86],"suggested":[87],"Out-of-Distribution":[88],"(OOD)":[89],"detection":[90,126],"as":[91],"an":[92],"additional":[93],"reliability":[94],"measure":[95],"when":[96],"deploying":[97],"real":[101],"world.":[102],"How":[103],"these":[104],"mechanisms":[105],"benefit":[106],"entire":[111],"system,":[112],"however,":[113],"not":[115],"yet":[116],"fully":[117],"understood.":[118],"work,":[121],"we":[122,175],"study":[123],"how":[124,182,197],"OOD":[125,178,192],"contributes":[127],"by":[134],"describing":[135],"challenges":[137],"involved":[138],"with":[139],"detecting":[140],"unknown":[141,148,211],"situations.":[142],"We":[143],"derive":[144],"several":[145],"definitions":[146],"events":[149,193],"explore":[151],"avenues":[153],"successful":[156],"argumentation,":[158],"building":[159],"on":[160,207],"recent":[161],"work":[162],"assurance":[165],"Machine":[167],"components.":[169],"series":[172],"experiments,":[174],"compare":[176],"different":[177],"detectors":[179],"show":[181],"difficult":[183],"distinguish":[187],"harmless":[188],"potentially":[190],"unsafe":[191],"practice,":[195],"standard":[198],"evaluation":[199],"schemes":[200],"can":[201],"lead":[202],"deceptive":[204],"conclusions,":[205],"depending":[206],"which":[208],"definition":[209],"applied.":[213]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":1}],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}