{"id":"https://openalex.org/W4381826884","doi":"https://doi.org/10.48550/arxiv.2306.12594","title":"State-wise Constrained Policy Optimization","display_name":"State-wise Constrained Policy Optimization","publication_year":2023,"publication_date":"2023-06-21","ids":{"openalex":"https://openalex.org/W4381826884","doi":"https://doi.org/10.48550/arxiv.2306.12594"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2306.12594","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.12594","pdf_url":"https://arxiv.org/pdf/2306.12594","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2306.12594","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5061774277","display_name":"Weiye Zhao","orcid":"https://orcid.org/0000-0002-8426-5238","observed_orcids":["https://orcid.org/0000-0002-8426-5238","https://orcid.org/0000-0002-7763-3119"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhao, Weiye","raw_affiliation_strings":["Robotics Institute Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Robotics Institute Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100403681","display_name":"Rui Chen","orcid":"https://orcid.org/0000-0003-1169-7678","observed_orcids":["https://orcid.org/0000-0003-1169-7678"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chen, Rui","raw_affiliation_strings":["Robotics Institute Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Robotics Institute Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041025969","display_name":"Yifan Sun","orcid":"https://orcid.org/0000-0003-3532-6521","observed_orcids":["https://orcid.org/0000-0003-3532-6521","https://orcid.org/0000-0003-2475-3843"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sun, Yifan","raw_affiliation_strings":["Robotics Institute Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Robotics Institute Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081793768","display_name":"Tianhao Wei","orcid":"https://orcid.org/0000-0003-2505-4585","observed_orcids":["https://orcid.org/0000-0003-2505-4585"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wei, Tianhao","raw_affiliation_strings":["Robotics Institute Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Robotics Institute Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5040156274","display_name":"Changliu Liu","orcid":"https://orcid.org/0000-0002-3767-5517","observed_orcids":["https://orcid.org/0000-0002-3767-5517","https://orcid.org/0000-0003-4923-3633"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Liu, Changliu","raw_affiliation_strings":["Robotics Institute Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Robotics Institute Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I74973139"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":16,"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":1.0,"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":1.0,"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/T12794","display_name":"Adaptive Dynamic Programming Control","score":0.0,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.0,"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/constrained-policy-optimization","display_name":"constrained policy optimization","score":0.7120000123977661},{"id":"https://openalex.org/keywords/safe-reinforcement-learning","display_name":"safe reinforcement learning","score":0.8489999771118164},{"id":"https://openalex.org/keywords/constrained-markov-decision-process","display_name":"constrained Markov decision process","score":0.8309999704360962},{"id":"https://openalex.org/keywords/robot-locomotion","display_name":"robot locomotion","score":0.8420000076293945},{"id":"https://openalex.org/keywords/robotic-manipulation","display_name":"robotic manipulation","score":0.6639999747276306},{"id":"https://openalex.org/keywords/autonomous-driving","display_name":"autonomous driving","score":0.8550000190734863},{"id":"https://openalex.org/keywords/neural-network-policies","display_name":"neural network policy","score":0.8629999756813049}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.851708710193634},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.7609965205192566},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6504160761833191},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.6146384477615356},{"id":"https://openalex.org/C44616089","wikidata":"https://www.wikidata.org/wiki/Q30158686","display_name":"Constraint satisfaction","level":3,"score":0.608148455619812},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.5897666215896606},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5564047694206238},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.5447396636009216},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.5380176305770874},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.5353954434394836},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4959035813808441},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.483639121055603},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.4254136085510254},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.3953101634979248},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.37691178917884827},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.17624062299728394},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.13789793848991394},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.12663519382476807},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"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},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.0},{"id":"https://openalex.org/C78519656","wikidata":"https://www.wikidata.org/wiki/Q101333","display_name":"Mechanical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2306.12594","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.12594","pdf_url":"https://arxiv.org/pdf/2306.12594","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2306.12594","doi":"https://doi.org/10.48550/arxiv.2306.12594","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2306.12594","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"article-journal"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2306.12594","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.12594","pdf_url":"https://arxiv.org/pdf/2306.12594","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4381826884.pdf","grobid_xml":"https://content.openalex.org/works/W4381826884.grobid-xml"},"referenced_works_count":11,"referenced_works":["https://openalex.org/W2788014517","https://openalex.org/W2804791273","https://openalex.org/W2911618937","https://openalex.org/W2994712737","https://openalex.org/W2998619042","https://openalex.org/W3118621910","https://openalex.org/W3130843035","https://openalex.org/W3164473225","https://openalex.org/W4283216217","https://openalex.org/W4382317755","https://openalex.org/W4402402057"],"related_works":["https://openalex.org/W4403883613","https://openalex.org/W4385767976","https://openalex.org/W4386231407","https://openalex.org/W4416590955","https://openalex.org/W2616964725","https://openalex.org/W4387185458","https://openalex.org/W4362650445","https://openalex.org/W2997273686","https://openalex.org/W3164473225","https://openalex.org/W7164861735"],"abstract_inverted_index":{"Reinforcement":[0],"Learning":[1],"(RL)":[2],"algorithms":[3,48],"have":[4],"shown":[5],"tremendous":[6],"success":[7],"in":[8,92,158],"simulation":[9],"environments,":[10],"but":[11],"their":[12],"application":[13],"to":[14],"real-world":[15],"problems":[16],"faces":[17],"significant":[18],"challenges,":[19],"with":[20],"safety":[21,110,142],"being":[22],"a":[23,138],"major":[24],"concern.":[25],"In":[26,94],"particular,":[27,95],"enforcing":[28],"state-wise":[29,61,81,89,141,156],"constraints":[30,157],"is":[31,112],"essential":[32],"for":[33,80,88,128],"many":[34],"challenging":[35],"tasks":[36],"such":[37],"as":[38],"autonomous":[39],"driving":[40],"and":[41,105,153],"robot":[42,130],"manipulation.":[43],"However,":[44],"existing":[45,151],"safe":[46],"RL":[47],"under":[49,114],"the":[50,74,98,108,118,134],"framework":[51,99],"of":[52,100,120,140],"Constrained":[53,70],"Markov":[54,102],"Decision":[55,103],"Process":[56],"(CMDP)":[57],"do":[58],"not":[59],"consider":[60],"constraints.":[62,143],"To":[63],"address":[64],"this":[65],"gap,":[66],"we":[67,96],"propose":[68],"State-wise":[69],"Policy":[71],"Optimization":[72],"(SCPO),":[73],"first":[75],"general-purpose":[76],"policy":[77],"search":[78],"algorithm":[79],"constrained":[82],"reinforcement":[83],"learning.":[84],"SCPO":[85,148],"provides":[86],"guarantees":[87],"constraint":[90],"satisfaction":[91],"expectation.":[93],"introduce":[97],"Maximum":[101],"Process,":[104],"prove":[106],"that":[107,147],"worst-case":[109],"violation":[111],"bounded":[113],"SCPO.":[115],"We":[116],"demonstrate":[117],"effectiveness":[119],"our":[121],"approach":[122],"on":[123],"training":[124],"neural":[125],"network":[126],"policies":[127],"extensive":[129],"locomotion":[131],"tasks,":[132],"where":[133],"agent":[135],"must":[136],"satisfy":[137],"variety":[139],"Our":[144],"results":[145],"show":[146],"significantly":[149],"outperforms":[150],"methods":[152],"can":[154],"handle":[155],"high-dimensional":[159],"robotics":[160],"tasks.":[161]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":1}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2023-06-24T00:00:00"}