{"id":"https://openalex.org/W3179916441","doi":"https://doi.org/10.1162/isal_a_00449","title":"Safer Reinforcement Learning through Transferable Instinct Networks","display_name":"Safer Reinforcement Learning through Transferable Instinct Networks","publication_year":2021,"publication_date":"2021-01-01","ids":{"openalex":"https://openalex.org/W3179916441","doi":"https://doi.org/10.1162/isal_a_00449","mag":"3179916441"},"language":"en","primary_location":{"id":"doi:10.1162/isal_a_00449","is_oa":false,"landing_page_url":"https://doi.org/10.1162/isal_a_00449","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The 2021 Conference on Artificial Life","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2107.06686","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5089907783","display_name":"Djordje Grbic","orcid":"https://orcid.org/0000-0002-4388-2539"},"institutions":[{"id":"https://openalex.org/I83467386","display_name":"IT University of Copenhagen","ror":"https://ror.org/02309jg23","country_code":"DK","type":"education","lineage":["https://openalex.org/I83467386"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Djordje Grbic","raw_affiliation_strings":["IT Univ. of Copenhagen, Copenhagen"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IT Univ. of Copenhagen, Copenhagen","institution_ids":["https://openalex.org/I83467386"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5020511097","display_name":"Sebastian Risi","orcid":"https://orcid.org/0000-0003-3607-8400"},"institutions":[{"id":"https://openalex.org/I83467386","display_name":"IT University of Copenhagen","ror":"https://ror.org/02309jg23","country_code":"DK","type":"education","lineage":["https://openalex.org/I83467386"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Sebastian Risi","raw_affiliation_strings":["IT-UNIVERSITY OF COPENHAGEN"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IT-UNIVERSITY OF COPENHAGEN","institution_ids":["https://openalex.org/I83467386"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I83467386"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9998000264167786,"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":0.9998000264167786,"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":0.9995999932289124,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9902999997138977,"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/safer","display_name":"SAFER","score":0.9048334360122681},{"id":"https://openalex.org/keywords/instinct","display_name":"Instinct","score":0.8914984464645386},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7935639023780823},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.7321016788482666},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6057949066162109},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.5111368298530579},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.4902544319629669},{"id":"https://openalex.org/keywords/risk-analysis","display_name":"Risk analysis (engineering)","score":0.4605506956577301},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.4337673783302307},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4318680167198181},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.4296889901161194},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.23184478282928467},{"id":"https://openalex.org/keywords/business","display_name":"Business","score":0.1235816478729248},{"id":"https://openalex.org/keywords/law","display_name":"Law","score":0.08161446452140808},{"id":"https://openalex.org/keywords/political-science","display_name":"Political science","score":0.07880091667175293},{"id":"https://openalex.org/keywords/systems-engineering","display_name":"Systems engineering","score":0.07527947425842285}],"concepts":[{"id":"https://openalex.org/C2776654903","wikidata":"https://www.wikidata.org/wiki/Q2601463","display_name":"SAFER","level":2,"score":0.9048334360122681},{"id":"https://openalex.org/C3232509","wikidata":"https://www.wikidata.org/wiki/Q18237","display_name":"Instinct","level":2,"score":0.8914984464645386},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7935639023780823},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.7321016788482666},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6057949066162109},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.5111368298530579},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.4902544319629669},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.4605506956577301},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.4337673783302307},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4318680167198181},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.4296889901161194},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.23184478282928467},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.1235816478729248},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.08161446452140808},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.07880091667175293},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.07527947425842285},{"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/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"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/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":5,"locations":[{"id":"doi:10.1162/isal_a_00449","is_oa":false,"landing_page_url":"https://doi.org/10.1162/isal_a_00449","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The 2021 Conference on Artificial Life","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2107.06686","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2107.06686","pdf_url":"https://arxiv.org/pdf/2107.06686","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"mag:3179916441","is_oa":true,"landing_page_url":"http://export.arxiv.org/pdf/2107.06686","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":"pmh:oai:pure.atira.dk:publications/1e472bea-7f01-4e7a-a4ab-b9ac2c193e86","is_oa":false,"landing_page_url":"https://pure.itu.dk/portal/da/publications/1e472bea-7f01-4e7a-a4ab-b9ac2c193e86","pdf_url":null,"source":{"id":"https://openalex.org/S4377196680","display_name":"IT University Of Copenhagen (IT University of Copenhagen)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I83467386","host_organization_name":"IT University of Copenhagen","host_organization_lineage":["https://openalex.org/I83467386"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Grbic, D & Risi, S 2021, 'Safer Reinforcement Learning through Transferable Instinct Networks', Artificial Life Conference Proceedings , no. 2021.","raw_type":"info:eu-repo/semantics/publishedVersion"},{"id":"doi:10.48550/arxiv.2107.06686","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2107.06686","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2107.06686","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2107.06686","pdf_url":"https://arxiv.org/pdf/2107.06686","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","score":0.4399999976158142,"display_name":"Peace, Justice and strong institutions"}],"awards":[{"id":"https://openalex.org/G4713059963","display_name":null,"funder_award_id":"FA8750","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"},{"id":"https://openalex.org/G5969032176","display_name":null,"funder_award_id":"FA8750-18-C-0103","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"}],"funders":[{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3179916441.pdf","grobid_xml":"https://content.openalex.org/works/W3179916441.grobid-xml"},"referenced_works_count":33,"referenced_works":["https://openalex.org/W1518931405","https://openalex.org/W1522301498","https://openalex.org/W1677182931","https://openalex.org/W1757796397","https://openalex.org/W1845972764","https://openalex.org/W1994616650","https://openalex.org/W2009797711","https://openalex.org/W2096952131","https://openalex.org/W2119717200","https://openalex.org/W2156737235","https://openalex.org/W2257979135","https://openalex.org/W2287376978","https://openalex.org/W2288565641","https://openalex.org/W2555811267","https://openalex.org/W2580175322","https://openalex.org/W2736601468","https://openalex.org/W2749928749","https://openalex.org/W2760552731","https://openalex.org/W2889696653","https://openalex.org/W2890803796","https://openalex.org/W2898621204","https://openalex.org/W2922874720","https://openalex.org/W2949608212","https://openalex.org/W2955766841","https://openalex.org/W2962867954","https://openalex.org/W2962991582","https://openalex.org/W2963575966","https://openalex.org/W2976866784","https://openalex.org/W2982316857","https://openalex.org/W2994180726","https://openalex.org/W3043774672","https://openalex.org/W3095135648","https://openalex.org/W3096954237"],"related_works":["https://openalex.org/W3184436868","https://openalex.org/W2885208901","https://openalex.org/W1541535141","https://openalex.org/W2736629007","https://openalex.org/W2952991114","https://openalex.org/W2574075983","https://openalex.org/W2330409864","https://openalex.org/W2774719390","https://openalex.org/W3166157537","https://openalex.org/W2271253566","https://openalex.org/W2342846537","https://openalex.org/W2788162021","https://openalex.org/W2592811292","https://openalex.org/W2597247754","https://openalex.org/W2784773199","https://openalex.org/W3184983782","https://openalex.org/W2952614427","https://openalex.org/W74200468","https://openalex.org/W3008890934","https://openalex.org/W2602543392"],"abstract_inverted_index":{"Random":[0],"exploration":[1],"is":[2,35,58,123,140,148,164],"one":[3,36],"of":[4,37,51,183],"the":[5,38,105,131,170],"main":[6,106],"mechanisms":[7],"through":[8],"which":[9,64,83,158,176],"reinforcement":[10],"learning":[11,26,34,132,159],"(RL)":[12],"finds":[13],"well-performing":[14],"policies.":[15],"However,":[16],"it":[17,147,177],"can":[18,45,66,87,103],"lead":[19,88],"to":[20,59,91,125,150,155],"undesirable":[21,127],"or":[22],"catastrophic":[23],"outcomes":[24],"when":[25],"online":[27],"in":[28,63,72,157,169,175],"safety-critical":[29],"environments.":[30],"In":[31,114],"fact,":[32],"safe":[33,149],"major":[39],"obstacles":[40],"towards":[41],"real-world":[42],"agents":[43,54],"that":[44,53],"learn":[46],"during":[47,186],"deployment.":[48],"One":[49],"way":[50],"ensuring":[52],"respect":[55],"hard":[56],"limitations":[57],"explicitly":[60],"configure":[61],"boundaries":[62],"they":[65],"operate.":[67],"While":[68],"this":[69],"might":[70],"work":[71],"some":[73],"cases,":[74],"we":[75,95],"do":[76],"not":[77],"always":[78],"have":[79],"clear":[80],"a-priori":[81],"information":[82],"states":[84],"and":[85,108,153],"actions":[86],"dangerously":[89],"close":[90],"hazardous":[92],"states.":[93],"Here,":[94],"present":[96],"an":[97,100,120],"approach":[98,192],"where":[99,146],"additional":[101],"policy":[102,107,133],"override":[104],"offer":[109],"a":[110,143,160,179,189],"safer":[111],"alternative":[112],"action.":[113],"our":[115],"instinct-regulated":[116],"RL":[117,191],"(IR^2L)":[118],"approach,":[119],"\"instinctual\"":[121],"network":[122,139],"trained":[124],"recognize":[126],"situations,":[128],"while":[129,193],"guarding":[130],"against":[134],"entering":[135],"them.":[136],"The":[137],"instinct":[138],"pre-trained":[141],"on":[142],"single":[144],"task":[145,162,196],"make":[151],"mistakes,":[152],"transferred":[154],"environments":[156],"new":[161],"safely":[163],"critical.":[165],"We":[166],"demonstrate":[167],"IR^2L":[168],"OpenAI":[171],"Safety":[172],"gym":[173],"domain,":[174],"receives":[178],"significantly":[180],"lower":[181],"number":[182],"safety":[184],"violations":[185],"training":[187],"than":[188],"baseline":[190],"reaching":[194],"similar":[195],"performance.":[197]},"counts_by_year":[{"year":2021,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}