{"id":"https://openalex.org/W182439313","doi":null,"title":"Avoiding failure states during reinforcement learning","display_name":"Avoiding failure states during reinforcement learning","publication_year":2011,"publication_date":"2011-02-24","ids":{"openalex":"https://openalex.org/W182439313","mag":"182439313"},"language":"en","primary_location":{"id":"pmh:oai:tudelft.nl:uuid:1f03c580-9fd5-4807-87b5-d70890e05ff6","doi":null,"is_oa":false,"landing_page_url":"http://resolver.tudelft.nl/uuid:1f03c580-9fd5-4807-87b5-d70890e05ff6","pdf_url":null,"source":{"id":"https://openalex.org/S4306400906","display_name":"Research Repository (Delft University of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I98358874","host_organization_name":"Delft University of Technology","host_organization_lineage":["https://openalex.org/I98358874"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"master thesis"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"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/A5070955240","display_name":"M.D.M. Van Diepen","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"M.D.M. Van Diepen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5070955240"],"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9994999766349792,"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":0.9994999766349792,"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/T10879","display_name":"Robotic Locomotion and Control","score":0.00039999998989515007,"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/T11023","display_name":"Prosthetics and Rehabilitation Robotics","score":0.0,"subfield":{"id":"https://openalex.org/subfields/2204","display_name":"Biomedical 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/bipedal-humanoid-robot","display_name":"bipedal humanoid robot","score":0.7310000061988831},{"id":"https://openalex.org/keywords/learning-to-walk","display_name":"learning to walk","score":0.7599999904632568},{"id":"https://openalex.org/keywords/fall-prevention","display_name":"fall prevention","score":0.6510000228881836},{"id":"https://openalex.org/keywords/failure-states","display_name":"failure states","score":0.8009999990463257},{"id":"https://openalex.org/keywords/action-selection","display_name":"action selection","score":0.8130000233650208},{"id":"https://openalex.org/keywords/pre-learning","display_name":"pre-learning","score":0.8040000200271606},{"id":"https://openalex.org/keywords/inverted-pendulum","display_name":"inverted pendulum","score":0.6309999823570251},{"id":"https://openalex.org/keywords/robot-learning","display_name":"robot learning","score":0.4020000100135803}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.7778995633125305},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6831324100494385},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.600254476070404},{"id":"https://openalex.org/C166109690","wikidata":"https://www.wikidata.org/wiki/Q4677422","display_name":"Action selection","level":3,"score":0.5940532088279724},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.5678418278694153},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.5536707043647766},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.546053409576416},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.5119422674179077},{"id":"https://openalex.org/C183759332","wikidata":"https://www.wikidata.org/wiki/Q343680","display_name":"Action learning","level":4,"score":0.49039673805236816},{"id":"https://openalex.org/C188888258","wikidata":"https://www.wikidata.org/wiki/Q7353390","display_name":"Robot learning","level":4,"score":0.47350436449050903},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.471964955329895},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4500208795070648},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.4342963397502899},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4141586720943451},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4112108647823334},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.34633228182792664},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.16849705576896667},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.1620534062385559},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.14196112751960754},{"id":"https://openalex.org/C51672120","wikidata":"https://www.wikidata.org/wiki/Q303446","display_name":"Cooperative learning","level":3,"score":0.1202704906463623},{"id":"https://openalex.org/C19966478","wikidata":"https://www.wikidata.org/wiki/Q4810574","display_name":"Mobile robot","level":3,"score":0.10132443904876709},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C145420912","wikidata":"https://www.wikidata.org/wiki/Q853077","display_name":"Mathematics education","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","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/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C66938386","wikidata":"https://www.wikidata.org/wiki/Q633538","display_name":"Structural engineering","level":1,"score":0.0},{"id":"https://openalex.org/C88610354","wikidata":"https://www.wikidata.org/wiki/Q1813494","display_name":"Teaching method","level":2,"score":0.0},{"id":"https://openalex.org/C169760540","wikidata":"https://www.wikidata.org/wiki/Q207011","display_name":"Neuroscience","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:tudelft.nl:uuid:1f03c580-9fd5-4807-87b5-d70890e05ff6","doi":null,"is_oa":false,"landing_page_url":"http://resolver.tudelft.nl/uuid:1f03c580-9fd5-4807-87b5-d70890e05ff6","pdf_url":null,"source":{"id":"https://openalex.org/S4306400906","display_name":"Research Repository (Delft University of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I98358874","host_organization_name":"Delft University of Technology","host_organization_lineage":["https://openalex.org/I98358874"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"master thesis"},{"id":"mag:182439313","doi":null,"is_oa":false,"landing_page_url":"http://repository.tudelft.nl/assets/uuid:1f03c580-9fd5-4807-87b5-d70890e05ff6/Thesis_M_van_Diepen.pdf","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":4,"referenced_works":["https://openalex.org/W183625566","https://openalex.org/W1980035368","https://openalex.org/W1984120922","https://openalex.org/W2121517924"],"related_works":["https://openalex.org/W2324948922","https://openalex.org/W7152665517","https://openalex.org/W585183151","https://openalex.org/W7161684719","https://openalex.org/W7134824097","https://openalex.org/W2292071423","https://openalex.org/W2771101156","https://openalex.org/W7096761421","https://openalex.org/W3102644508","https://openalex.org/W2808262291"],"abstract_inverted_index":{"The":[0,153,177,266,321],"Delft":[1],"Biorobotics":[2],"Laboratory":[3],"develops":[4],"bipedal":[5],"humanoid":[6],"robots.":[7],"One":[8],"of":[9,46,91,150,155,165,183,228,251,313,328],"these":[10],"robots,":[11],"called":[12],"LEO,":[13],"is":[14,53,73,126,210,271,286],"designed":[15],"to":[16,18,38,69,76,161,224,273,288,297],"learn":[17],"walk":[19],"using":[20,309,347],"reinforcement":[21],"learning.":[22],"During":[23],"learning,":[24,94],"LEO":[25],"will":[26],"make":[27,77],"mistakes":[28,32],"and":[29,235],"fall.":[30],"These":[31],"can":[33,67,334,355],"cause":[34],"serious":[35],"dam-":[36],"age":[37],"the":[39,47,55,63,89,97,102,122,130,142,159,163,181,197,202,226,239,275,281,290,299,303,326,348],"system":[40],"but":[41,214],"are":[42],"an":[43,124],"integral":[44],"part":[45],"learning":[48,103,198,231,240],"process.":[49],"A":[50],"likely":[51],"solution":[52],"punishing":[54,62],"robot":[56,64,71],"more":[57,284],"severely":[58],"for":[59,114,133,302,351,358],"falling.":[60],"However,":[61],"too":[65,74],"rigorously":[66],"lead":[68],"a":[70,78,111,146,166,211,249,310,336],"that":[72,87,135,361],"cautious":[75],"step.":[79],"In":[80],"this":[81,189],"research,":[82],"three":[83],"methods":[84],"were":[85],"tested":[86,172],"reduce":[88,225],"amount":[90,182,227,327],"falls":[92,184,229,329],"during":[93,119,230],"without":[95],"restricting":[96],"possible":[98],"solutions":[99],"or":[100],"increasing":[101],"time.":[104,199],"We":[105],"introduce":[106],"Threshold":[107],"Restricted":[108],"Learning":[109],"(TRL),":[110],"new":[112,256],"method":[113,316],"action":[115,125,208],"se-":[116],"lection":[117],"that,":[118],"exploration,":[120],"makes":[121],"probability":[123,149],"chosen":[127],"dependent":[128],"on":[129],"expected":[131,139,259,276,291],"reward":[132],"taking":[134],"action.":[136],"Actions":[137],"with":[138,219,232,268,283],"rewards":[140,260,360],"below":[141],"set":[143],"threshold":[144],"have":[145,335],"signifcant":[147],"reduced":[148],"being":[151],"chosen.":[152],"concept":[154],"TRL":[156,170,187,193],"developed":[157],"from":[158],"desire":[160],"optimize":[162],"use":[164,319,338],"pre-learned":[167],"solution.":[168],"Hence,":[169],"was":[171,191,205,222,330],"after":[173],"pre-learning":[174,201,220],"in":[175,180,188,325],"simulation.":[176],"largest":[178,203],"reduction":[179,204,250,324],"achieved":[185,323],"by":[186],"research":[190],"50%":[192],"did":[194,236],"not":[195,237,318],"increase":[196,238],"Without":[200,243],"7%.":[206],"Softmax":[207,245],"selection":[209,216],"well":[212],"known":[213],"underused":[215],"method.":[217],"Combined":[218],"it":[221,354],"able":[223],"approximately":[233,252,331],"80%":[234],"time":[241],"either.":[242],"pre-learning,":[244],"could":[246],"still":[247],"achieve":[248],"20%.":[253,332],"Sarsa2Q,":[254],"another":[255],"method,":[257],"stores":[258],"at":[261],"different":[262],"levels":[263],"op":[264],"generalization.":[265,314],"level":[267,282,312],"little":[269],"generalization":[270,285,350],"used":[272,287,357],"store":[274,289],"total":[277,292],"positive":[278],"reward,":[279],"while":[280],"negative":[293,352],"reward.":[294],"Sarsa2Q":[295,333],"learned":[296],"avoid":[298],"failure":[300,342],"states":[301],"inverted":[304],"pendulum":[305],"problem":[306],"faster":[307],"than":[308,339,345],"single":[311],"This":[315],"does":[317],"pre-learning.":[320],"highest":[322],"broader":[337],"just":[340],"avoiding":[341],"states.":[343],"Rather":[344],"only":[346],"coarse":[349],"rewards,":[353],"be":[356],"all":[359],"need":[362],"less":[363],"precision.":[364]},"counts_by_year":[],"updated_date":"2026-08-18T07:49:30.821534","created_date":"2016-06-24T00:00:00"}