{"id":"https://openalex.org/W4312347470","doi":"https://doi.org/10.1109/iros47612.2022.9981831","title":"Safety Correction from Baseline: Towards the Risk-aware Policy in Robotics via Dual-agent Reinforcement Learning","display_name":"Safety Correction from Baseline: Towards the Risk-aware Policy in Robotics via Dual-agent Reinforcement Learning","publication_year":2022,"publication_date":"2022-10-23","ids":{"openalex":"https://openalex.org/W4312347470","doi":"https://doi.org/10.1109/iros47612.2022.9981831"},"language":"en","primary_location":{"id":"doi:10.1109/iros47612.2022.9981831","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros47612.2022.9981831","pdf_url":null,"source":{"id":"https://openalex.org/S4363607704","display_name":"2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","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/A5045272072","display_name":"Linrui Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114105","display_name":"Tsinghua\u2013Berkeley Shenzhen Institute","ror":"https://ror.org/02hhwwz98","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210114105","https://openalex.org/I95457486","https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Linrui Zhang","raw_affiliation_strings":["Tsinghua Shenzhen International Graduate School,Center for Intelligent Control and Telescience,Shenzhen,China,518055"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua Shenzhen International Graduate School,Center for Intelligent Control and Telescience,Shenzhen,China,518055","institution_ids":["https://openalex.org/I4210114105"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028350523","display_name":"Zichen Yan","orcid":"https://orcid.org/0009-0008-1755-9309"},"institutions":[{"id":"https://openalex.org/I4210114105","display_name":"Tsinghua\u2013Berkeley Shenzhen Institute","ror":"https://ror.org/02hhwwz98","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210114105","https://openalex.org/I95457486","https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zichen Yan","raw_affiliation_strings":["Tsinghua Shenzhen International Graduate School,Center for Intelligent Control and Telescience,Shenzhen,China,518055"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua Shenzhen International Graduate School,Center for Intelligent Control and Telescience,Shenzhen,China,518055","institution_ids":["https://openalex.org/I4210114105"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100333320","display_name":"Li Shen","orcid":"https://orcid.org/0000-0002-5443-0503"},"institutions":[{"id":"https://openalex.org/I4210103986","display_name":"Jingdong (China)","ror":"https://ror.org/01dkjkq64","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210103986"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li Shen","raw_affiliation_strings":["JD Explore Academy,Beijing,China,100176"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"JD Explore Academy,Beijing,China,100176","institution_ids":["https://openalex.org/I4210103986"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036039550","display_name":"Shoujie Li","orcid":"https://orcid.org/0000-0002-5561-3379"},"institutions":[{"id":"https://openalex.org/I4210114105","display_name":"Tsinghua\u2013Berkeley Shenzhen Institute","ror":"https://ror.org/02hhwwz98","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210114105","https://openalex.org/I95457486","https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shoujie Li","raw_affiliation_strings":["Tsinghua Shenzhen International Graduate School,Center for Intelligent Control and Telescience,Shenzhen,China,518055"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua Shenzhen International Graduate School,Center for Intelligent Control and Telescience,Shenzhen,China,518055","institution_ids":["https://openalex.org/I4210114105"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100737125","display_name":"Xueqian Wang","orcid":"https://orcid.org/0000-0003-3542-0593"},"institutions":[{"id":"https://openalex.org/I4210114105","display_name":"Tsinghua\u2013Berkeley Shenzhen Institute","ror":"https://ror.org/02hhwwz98","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210114105","https://openalex.org/I95457486","https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xueqian Wang","raw_affiliation_strings":["Tsinghua Shenzhen International Graduate School,Center for Intelligent Control and Telescience,Shenzhen,China,518055"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua Shenzhen International Graduate School,Center for Intelligent Control and Telescience,Shenzhen,China,518055","institution_ids":["https://openalex.org/I4210114105"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5074103823","display_name":"Dacheng Tao","orcid":"https://orcid.org/0000-0001-7225-5449"},"institutions":[{"id":"https://openalex.org/I4210103986","display_name":"Jingdong (China)","ror":"https://ror.org/01dkjkq64","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210103986"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dacheng Tao","raw_affiliation_strings":["JD Explore Academy,Beijing,China,100176"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"JD Explore Academy,Beijing,China,100176","institution_ids":["https://openalex.org/I4210103986"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.3114,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.52409223,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":96},"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":0.9991999864578247,"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":0.9991999864578247,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.9959999918937683,"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/T11099","display_name":"Autonomous Vehicle Technology and Safety","score":0.9904999732971191,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive 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":0.8875836133956909},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.8294702768325806},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.73982834815979},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.7209175229072571},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.624193549156189},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5160993933677673},{"id":"https://openalex.org/keywords/dual","display_name":"Dual (grammatical number)","score":0.5139064788818359},{"id":"https://openalex.org/keywords/robotics","display_name":"Robotics","score":0.49840402603149414},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.4776393473148346},{"id":"https://openalex.org/keywords/controller","display_name":"Controller (irrigation)","score":0.4705168306827545},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.41694965958595276},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.07928705215454102}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8875836133956909},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.8294702768325806},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.73982834815979},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.7209175229072571},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.624193549156189},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5160993933677673},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.5139064788818359},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.49840402603149414},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.4776393473148346},{"id":"https://openalex.org/C203479927","wikidata":"https://www.wikidata.org/wiki/Q5165939","display_name":"Controller (irrigation)","level":2,"score":0.4705168306827545},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.41694965958595276},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.07928705215454102},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0},{"id":"https://openalex.org/C124952713","wikidata":"https://www.wikidata.org/wiki/Q8242","display_name":"Literature","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C111368507","wikidata":"https://www.wikidata.org/wiki/Q43518","display_name":"Oceanography","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iros47612.2022.9981831","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros47612.2022.9981831","pdf_url":null,"source":{"id":"https://openalex.org/S4363607704","display_name":"2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","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":38,"referenced_works":["https://openalex.org/W2340985688","https://openalex.org/W2462906003","https://openalex.org/W2560674852","https://openalex.org/W2741122588","https://openalex.org/W2747329762","https://openalex.org/W2784465508","https://openalex.org/W2788862220","https://openalex.org/W2913300629","https://openalex.org/W2963411833","https://openalex.org/W2992833799","https://openalex.org/W2994967560","https://openalex.org/W2998396902","https://openalex.org/W3011338904","https://openalex.org/W3016507688","https://openalex.org/W3129808907","https://openalex.org/W3174069331","https://openalex.org/W3184308941","https://openalex.org/W3186449011","https://openalex.org/W3187709438","https://openalex.org/W3189576409","https://openalex.org/W3195968524","https://openalex.org/W4206497039","https://openalex.org/W4287370827","https://openalex.org/W4292230442","https://openalex.org/W4300799055","https://openalex.org/W6677959772","https://openalex.org/W6684205842","https://openalex.org/W6718836005","https://openalex.org/W6737893269","https://openalex.org/W6740801417","https://openalex.org/W6747790125","https://openalex.org/W6748314335","https://openalex.org/W6748839928","https://openalex.org/W6751285671","https://openalex.org/W6751725685","https://openalex.org/W6758924645","https://openalex.org/W6768977950","https://openalex.org/W6797010411"],"related_works":["https://openalex.org/W2383111961","https://openalex.org/W2365952365","https://openalex.org/W2352448290","https://openalex.org/W2380820513","https://openalex.org/W2913146933","https://openalex.org/W2372385138","https://openalex.org/W4296359239","https://openalex.org/W2101155126","https://openalex.org/W2043093291","https://openalex.org/W2363545964"],"abstract_inverted_index":{"Learning":[0],"a":[1,49,57,60,64,143,154,168],"risk-aware":[2],"policy":[3,28,116,135],"is":[4,81,92],"essential":[5],"but":[6],"rather":[7],"challenging":[8],"in":[9,40],"unstructured":[10],"robotic":[11],"tasks.":[12],"Safe":[13],"reinforcement":[14,52],"learning":[15,53],"methods":[16],"open":[17],"up":[18],"new":[19],"possibilities":[20],"to":[21,33,120,130,141,213],"tackle":[22],"this":[23,45],"problem.":[24],"However,":[25],"the":[26,78,105,108,112,160,164,199],"conservative":[27],"updates":[29],"make":[30],"it":[31,91],"intractable":[32],"achieve":[34],"sufficient":[35],"exploration":[36,101],"and":[37,59,72,102,118,208,218],"desirable":[38],"performance":[39],"complex,":[41],"sample-expensive":[42],"environments.":[43],"In":[44,128],"paper,":[46],"we":[47],"propose":[48],"dual-agent":[50],"safe":[51,61,109,134,201],"strategy":[54],"consisting":[55],"of":[56,98],"baseline":[58,79,113,170],"agent.":[62,171],"Such":[63],"decoupled":[65],"framework":[66],"enables":[67],"high":[68],"flexibility,":[69],"data":[70],"efficiency":[71],"risk-awareness":[73],"for":[74,83,115],"RL-based":[75],"control.":[76],"Concretely,":[77],"agent":[80,110,114],"responsible":[82],"maximizing":[84],"rewards":[85],"under":[86],"standard":[87],"RL":[88,126,202],"settings.":[89],"Thus,":[90],"compatible":[93],"with":[94,211],"off-the-shelf":[95],"training":[96,131],"techniques":[97],"unconstrained":[99],"optimization,":[100],"exploitation.":[103],"On":[104],"other":[106],"hand,":[107],"mimics":[111],"improvement":[117],"learns":[119],"fulfill":[121],"safety":[122,215],"constraints":[123],"via":[124,153],"off-policy":[125],"tuning.":[127],"contrast":[129],"from":[132,191],"scratch,":[133],"correction":[136],"requires":[137],"significantly":[138],"fewer":[139],"interactions":[140],"obtain":[142],"near-optimal":[144],"policy.":[145],"The":[146,195],"dual":[147],"policies":[148],"can":[149,178],"be":[150],"optimized":[151],"synchronously":[152],"shared":[155],"replay":[156],"buffer,":[157],"or":[158,163],"leveraging":[159],"pre-trained":[161,192],"model":[162],"non-learning-based":[165],"controller":[166],"as":[167,185,187],"fixed":[169],"Experimental":[172],"results":[173],"show":[174],"that":[175],"our":[176],"approach":[177],"learn":[179],"feasible":[180],"skills":[181],"without":[182],"prior":[183],"knowledge":[184],"well":[186],"deriving":[188],"risk-averse":[189],"counterparts":[190],"unsafe":[193],"policies.":[194],"proposed":[196],"method":[197],"outperforms":[198],"state-of-the-art":[200],"algorithms":[203],"on":[204],"difficult":[205],"robot":[206],"locomotion":[207],"manipulation":[209],"tasks":[210],"respect":[212],"both":[214],"constraint":[216],"satisfaction":[217],"sample":[219],"efficiency.":[220]},"counts_by_year":[{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}