{"id":"https://openalex.org/W4414360302","doi":"https://doi.org/10.24963/ijcai.2025/696","title":"Rule-Guided Reinforcement Learning Policy Evaluation and Improvement","display_name":"Rule-Guided Reinforcement Learning Policy Evaluation and Improvement","publication_year":2025,"publication_date":"2025-09-01","ids":{"openalex":"https://openalex.org/W4414360302","doi":"https://doi.org/10.24963/ijcai.2025/696"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2025/696","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/696","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","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/A5078635542","display_name":"Martin Tappler","orcid":"https://orcid.org/0000-0002-4193-5609"},"institutions":[{"id":"https://openalex.org/I145847075","display_name":"TU Wien","ror":"https://ror.org/04d836q62","country_code":"AT","type":"education","lineage":["https://openalex.org/I145847075"]}],"countries":["AT"],"is_corresponding":true,"raw_author_name":"Martin Tappler","raw_affiliation_strings":["TU Wien"],"affiliations":[{"raw_affiliation_string":"TU Wien","institution_ids":["https://openalex.org/I145847075"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065480828","display_name":"Ignacio D. Lopez-Miguel","orcid":"https://orcid.org/0000-0002-8044-0385"},"institutions":[{"id":"https://openalex.org/I145847075","display_name":"TU Wien","ror":"https://ror.org/04d836q62","country_code":"AT","type":"education","lineage":["https://openalex.org/I145847075"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Ignacio D. Lopez-Miguel","raw_affiliation_strings":["TU Wien"],"affiliations":[{"raw_affiliation_string":"TU Wien","institution_ids":["https://openalex.org/I145847075"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051794439","display_name":"Sebastian Tschiatschek","orcid":"https://orcid.org/0000-0002-2592-0108"},"institutions":[{"id":"https://openalex.org/I129774422","display_name":"University of Vienna","ror":"https://ror.org/03prydq77","country_code":"AT","type":"education","lineage":["https://openalex.org/I129774422"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Sebastian Tschiatschek","raw_affiliation_strings":["University of Vienna"],"affiliations":[{"raw_affiliation_string":"University of Vienna","institution_ids":["https://openalex.org/I129774422"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5050836932","display_name":"Ezio Bartocci","orcid":"https://orcid.org/0000-0002-8004-6601"},"institutions":[{"id":"https://openalex.org/I145847075","display_name":"TU Wien","ror":"https://ror.org/04d836q62","country_code":"AT","type":"education","lineage":["https://openalex.org/I145847075"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Ezio Bartocci","raw_affiliation_strings":["TU Wien"],"affiliations":[{"raw_affiliation_string":"TU Wien","institution_ids":["https://openalex.org/I145847075"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5078635542"],"corresponding_institution_ids":["https://openalex.org/I145847075"],"apc_list":null,"apc_paid":null,"fwci":1.1617,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.82729576,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"6254","last_page":"6262"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T14011","display_name":"Elevator Systems and Control","score":0.5450000166893005,"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/T14011","display_name":"Elevator Systems and Control","score":0.5450000166893005,"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":0.8492000102996826},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.5595999956130981},{"id":"https://openalex.org/keywords/domain-knowledge","display_name":"Domain knowledge","score":0.477400004863739},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.3889999985694885},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.3481000065803528}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8492000102996826},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7376000285148621},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6261000037193298},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.5595999956130981},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.477400004863739},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4212999939918518},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.3889999985694885},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.3481000065803528},{"id":"https://openalex.org/C63882131","wikidata":"https://www.wikidata.org/wiki/Q17122954","display_name":"Strengths and weaknesses","level":2,"score":0.31459999084472656},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.30410000681877136},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2930000126361847}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.24963/ijcai.2025/696","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/696","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,73,150],"consider":[1],"the":[2,48,58,63,114,121,152],"challenging":[3],"problem":[4],"of":[5,86,141,154,167],"using":[6,66],"domain":[7,67,145],"knowledge":[8,68,146],"to":[9,80,82,90,101],"improve":[10,105],"deep":[11,35],"reinforcement":[12],"learning":[13],"policies.":[14],"To":[15],"this":[16],"end,":[17],"we":[18,61],"propose":[19],"LEGIBLE,":[20],"a":[21,25,34,39,138],"novel":[22],"approach,":[23],"following":[24],"multi-step":[26],"process,":[27],"which":[28,46,53,103],"starts":[29],"by":[30,129,134,157,165,175],"mining":[31],"rules":[32,44,65,100,123,182],"from":[33,77],"RL":[36,49,81,148,172],"policy,":[37,115],"constituting":[38],"partially":[40],"symbolic":[41],"representation.":[42],"These":[43,109],"describe":[45],"decisions":[47],"policy":[50,179],"makes":[51],"and":[52,124,143,174],"it":[54,117,160],"avoids":[55],"making.":[56],"In":[57],"second":[59],"step,":[60],"generalize":[62],"mined":[64],"expressed":[69],"as":[70],"metamorphic":[71,135],"relations.":[72],"adapt":[74],"these":[75,168],"relations":[76,136],"software":[78],"testing":[79],"specify":[83],"expected":[84],"changes":[85,91],"actions":[87],"in":[88,92,113,170],"response":[89],"observations.":[93],"The":[94],"third":[95],"step":[96],"is":[97],"evaluating":[98],"generalized":[99],"determine":[102],"generalizations":[104],"performance":[106,184],"when":[107],"enforced.":[108],"improvements":[110],"show":[111,151],"weaknesses":[112],"where":[116],"has":[118],"not":[119],"learned":[120],"general":[122],"thus":[125],"can":[126],"be":[127],"improved":[128],"rule":[130],"guidance.":[131],"LEGIBLE":[132],"supported":[133],"provides":[137],"principled":[139],"way":[140],"expressing":[142],"enforcing":[144],"about":[147],"environments.":[149],"efficacy":[153],"our":[155],"approach":[156],"demonstrating":[158],"that":[159,177],"effectively":[161],"finds":[162],"weaknesses,":[163,169],"accompanied":[164],"explanations":[166],"eleven":[171],"environments":[173],"showcasing":[176],"guiding":[178],"execution":[180],"with":[181],"improves":[183],"w.r.t.":[185],"gained":[186],"reward.":[187]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}