{"id":"https://openalex.org/W4415180223","doi":"https://doi.org/10.23919/ecc65951.2025.11186958","title":"Predictive Safety Shield for Dyna-Q Reinforcement Learning","display_name":"Predictive Safety Shield for Dyna-Q Reinforcement Learning","publication_year":2025,"publication_date":"2025-06-24","ids":{"openalex":"https://openalex.org/W4415180223","doi":"https://doi.org/10.23919/ecc65951.2025.11186958"},"language":"en","primary_location":{"id":"doi:10.23919/ecc65951.2025.11186958","is_oa":false,"landing_page_url":"https://doi.org/10.23919/ecc65951.2025.11186958","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 European Control Conference (ECC)","raw_type":"proceedings-article"},"type":"article","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/A5119998311","display_name":"Pin Jin","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Pin Jin","raw_affiliation_strings":["Institut Polytechnique de Paris,ENSTA Paris,Departement U2IS"],"affiliations":[{"raw_affiliation_string":"Institut Polytechnique de Paris,ENSTA Paris,Departement U2IS","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065905965","display_name":"Hanna Krasowski","orcid":"https://orcid.org/0000-0002-6730-3802"},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hanna Krasowski","raw_affiliation_strings":["University of California,Department of Electrical Engineering and Computer Science,Berkeley,USA"],"affiliations":[{"raw_affiliation_string":"University of California,Department of Electrical Engineering and Computer Science,Berkeley,USA","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5119665443","display_name":"Elena Vanneaux","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Elena Vanneaux","raw_affiliation_strings":["Institut Polytechnique de Paris,ENSTA Paris,Departement U2IS"],"affiliations":[{"raw_affiliation_string":"Institut Polytechnique de Paris,ENSTA Paris,Departement U2IS","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5119998311"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.39222822,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2173","last_page":"2179"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10876","display_name":"Fault Detection and Control Systems","score":0.6215999722480774,"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"}},"topics":[{"id":"https://openalex.org/T10876","display_name":"Fault Detection and Control Systems","score":0.6215999722480774,"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"}},{"id":"https://openalex.org/T13295","display_name":"Safety Systems Engineering in Autonomy","score":0.6104999780654907,"subfield":{"id":"https://openalex.org/subfields/2213","display_name":"Safety, Risk, Reliability and Quality"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11357","display_name":"Risk and Safety Analysis","score":0.5893999934196472,"subfield":{"id":"https://openalex.org/subfields/1804","display_name":"Statistics, Probability and Uncertainty"},"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.730400025844574},{"id":"https://openalex.org/keywords/shield","display_name":"Shield","score":0.6614999771118164},{"id":"https://openalex.org/keywords/shields","display_name":"Shields","score":0.590399980545044},{"id":"https://openalex.org/keywords/reinforcement","display_name":"Reinforcement","score":0.527899980545044},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.31610000133514404}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.730400025844574},{"id":"https://openalex.org/C138081364","wikidata":"https://www.wikidata.org/wiki/Q852013","display_name":"Shield","level":2,"score":0.6614999771118164},{"id":"https://openalex.org/C109589588","wikidata":"https://www.wikidata.org/wiki/Q2253638","display_name":"Shields","level":3,"score":0.590399980545044},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.527899980545044},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4659999907016754},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.3898000121116638},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.31610000133514404},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.29809999465942383},{"id":"https://openalex.org/C132835097","wikidata":"https://www.wikidata.org/wiki/Q7663745","display_name":"System safety","level":2,"score":0.2757999897003174},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.26330000162124634}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.23919/ecc65951.2025.11186958","is_oa":false,"landing_page_url":"https://doi.org/10.23919/ecc65951.2025.11186958","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 European Control Conference (ECC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W1491843047","https://openalex.org/W2162651021","https://openalex.org/W2790380176","https://openalex.org/W2963525569","https://openalex.org/W2963575966","https://openalex.org/W2992252352","https://openalex.org/W3011250830","https://openalex.org/W3093055384","https://openalex.org/W3107034996","https://openalex.org/W3114470638","https://openalex.org/W3135301907","https://openalex.org/W3159199672","https://openalex.org/W3188389400","https://openalex.org/W4214717370","https://openalex.org/W4280525627","https://openalex.org/W4285818823","https://openalex.org/W4312510909","https://openalex.org/W4372353038","https://openalex.org/W4382318502","https://openalex.org/W4387171936","https://openalex.org/W4391021830","https://openalex.org/W4391741307"],"related_works":[],"abstract_inverted_index":{"Obtaining":[0],"safety":[1,25,29,59,70,98],"guarantees":[2],"for":[3,13,61],"reinforcement":[4,20,63],"learning":[5,21,64],"is":[6,124],"a":[7,39,57,83],"major":[8],"challenge":[9],"to":[10,114,126],"achieve":[11,23],"applicability":[12],"real-world":[14],"tasks.":[15],"Safety":[16],"shields":[17,30],"extend":[18],"standard":[19],"and":[22,132],"hard":[24,97],"guarantees.":[26,99],"However,":[27],"existing":[28],"commonly":[31],"use":[32],"random":[33],"sampling":[34],"of":[35,48,86],"safe":[36,50,78,84],"actions":[37],"or":[38],"fixed":[40],"fallback":[41],"controller,":[42],"therefore":[43],"disregarding":[44],"future":[45],"performance":[46,94],"implications":[47],"different":[49],"actions.":[51],"In":[52],"this":[53],"work,":[54],"we":[55],"propose":[56],"predictive":[58],"shield":[60,71],"model-based":[62],"agents":[65],"in":[66],"discrete":[67],"space.":[68],"Our":[69,100],"updates":[72],"the":[73,87,116],"Q-function":[74],"locally":[75],"based":[76],"on":[77,102],"predictions,":[79],"which":[80],"originate":[81],"from":[82],"simulation":[85,131],"environment":[88],"model.":[89],"This":[90],"shielding":[91],"approach":[92,123],"improves":[93],"while":[95],"maintaining":[96],"experiments":[101],"gridworld":[103],"environments":[104],"demonstrate":[105],"that":[106,121],"even":[107],"short":[108],"prediction":[109],"horizons":[110],"can":[111],"be":[112],"sufficient":[113],"identify":[115],"optimal":[117],"path.":[118],"We":[119],"observe":[120],"our":[122],"robust":[125],"distribution":[127],"shifts,":[128],"e.g.,":[129],"between":[130],"reality,":[133],"without":[134],"requiring":[135],"additional":[136],"training.":[137]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-15T00:00:00"}
