{"id":"https://openalex.org/W7152959400","doi":"https://doi.org/10.48550/arxiv.2604.07875","title":"Learning over Forward-Invariant Policy Classes: Reinforcement Learning without Safety Concerns","display_name":"Learning over Forward-Invariant Policy Classes: Reinforcement Learning without Safety Concerns","publication_year":2026,"publication_date":"2026-04-09","ids":{"openalex":"https://openalex.org/W7152959400","doi":"https://doi.org/10.48550/arxiv.2604.07875"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.07875","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07875","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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.07875","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031927131","display_name":"C. T. Tsai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tsai, Chieh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043410562","display_name":"Muhammad Junayed Hasan Zahed","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zahed, Muhammad Junayed Hasan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133369210","display_name":"Salim Hariri","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hariri, Salim","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5085452242","display_name":"Hossein Rastgoftar","orcid":"https://orcid.org/0000-0001-9212-7161"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rastgoftar, Hossein","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.20999999344348907,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.20999999344348907,"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":0.20239999890327454,"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":0.1964000016450882,"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":0.8454999923706055},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.5884000062942505},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5654000043869019},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.5562000274658203},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.5060999989509583},{"id":"https://openalex.org/keywords/q-learning","display_name":"Q-learning","score":0.4343999922275543},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.4187000095844269},{"id":"https://openalex.org/keywords/foundation","display_name":"Foundation (evidence)","score":0.3652999997138977}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8454999923706055},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.5884000062942505},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5654000043869019},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.5562000274658203},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5465999841690063},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5060999989509583},{"id":"https://openalex.org/C188116033","wikidata":"https://www.wikidata.org/wiki/Q2664563","display_name":"Q-learning","level":3,"score":0.4343999922275543},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.4187000095844269},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4106000065803528},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.3652999997138977},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.350600004196167},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.34610000252723694},{"id":"https://openalex.org/C2779199153","wikidata":"https://www.wikidata.org/wiki/Q43965","display_name":"Quadcopter","level":2,"score":0.3449000120162964},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.3208000063896179},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.30820000171661377},{"id":"https://openalex.org/C17098449","wikidata":"https://www.wikidata.org/wiki/Q176814","display_name":"Partially observable Markov decision process","level":4,"score":0.2953000068664551},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.27730000019073486},{"id":"https://openalex.org/C132835097","wikidata":"https://www.wikidata.org/wiki/Q7663745","display_name":"System safety","level":2,"score":0.2766000032424927},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2630999982357025},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.26249998807907104}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.07875","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07875","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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.07875","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07875","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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"score":0.8180467486381531,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"This":[0],"paper":[1],"proposes":[2],"a":[3,20,45,57,66,78,87,123],"safe":[4,68,127],"reinforcement":[5],"learning":[6,128],"(RL)":[7],"framework":[8,85],"based":[9],"on":[10,28,86],"forward-invariance-induced":[11],"action-space":[12],"design.":[13],"The":[14,112],"control":[15],"problem":[16,90],"is":[17,35],"cast":[18],"as":[19],"Markov":[21],"decision":[22],"process,":[23],"but":[24],"instead":[25],"of":[26,65],"relying":[27],"runtime":[29],"shielding":[30],"or":[31],"penalty-based":[32],"constraints,":[33],"safety":[34,116],"embedded":[36],"directly":[37],"into":[38],"the":[39,72,84,97],"action":[40,48,54],"representation.":[41],"Specifically,":[42],"we":[43],"construct":[44],"finite":[46],"admissible":[47],"set":[49],"in":[50,129],"which":[51],"each":[52],"discrete":[53],"corresponds":[55],"to":[56],"stabilizing":[58],"feedback":[59],"law":[60],"that":[61,96],"preserves":[62],"forward":[63],"invariance":[64],"prescribed":[67],"state":[69],"set.":[70],"Consequently,":[71],"RL":[73],"agent":[74],"optimizes":[75],"policies":[76,109],"over":[77],"safe-by-construction":[79],"policy":[80,99],"class.":[81],"We":[82],"validate":[83],"quadcopter":[88],"hover-regulation":[89],"under":[91],"disturbance.":[92],"Simulation":[93],"results":[94],"show":[95],"learned":[98],"improves":[100],"closed-loop":[101],"performance":[102,119],"and":[103,121],"switching":[104],"efficiency,":[105],"while":[106],"all":[107],"evaluated":[108],"remain":[110],"safety-preserving.":[111],"proposed":[113],"formulation":[114],"decouples":[115],"assurance":[117],"from":[118],"optimization":[120],"provides":[122],"promising":[124],"foundation":[125],"for":[126],"nonlinear":[130],"systems.":[131]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-11T00:00:00"}