{"id":"https://openalex.org/W4415428247","doi":"https://doi.org/10.3233/faia251080","title":"Accelerating Transformers in Online RL","display_name":"Accelerating Transformers in Online RL","publication_year":2025,"publication_date":"2025-10-21","ids":{"openalex":"https://openalex.org/W4415428247","doi":"https://doi.org/10.3233/faia251080"},"language":null,"primary_location":{"id":"doi:10.3233/faia251080","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia251080","pdf_url":null,"source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},"type":"book-chapter","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.3233/faia251080","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120034034","display_name":"Daniil Zelezetsky","orcid":null},"institutions":[{"id":"https://openalex.org/I153845743","display_name":"Moscow Institute of Physics and Technology","ror":"https://ror.org/00v0z9322","country_code":"RU","type":"education","lineage":["https://openalex.org/I153845743"]}],"countries":["RU"],"is_corresponding":false,"raw_author_name":"Daniil Zelezetsky","raw_affiliation_strings":["MIPT, Dolgoprudny, Russia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MIPT, Dolgoprudny, Russia","institution_ids":["https://openalex.org/I153845743"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054791318","display_name":"Alexey K. Kovalev","orcid":"https://orcid.org/0000-0003-2180-0990"},"institutions":[{"id":"https://openalex.org/I153845743","display_name":"Moscow Institute of Physics and Technology","ror":"https://ror.org/00v0z9322","country_code":"RU","type":"education","lineage":["https://openalex.org/I153845743"]}],"countries":["RU"],"is_corresponding":false,"raw_author_name":"Alexey K. Kovalev","raw_affiliation_strings":["AIRI, Moscow, Russia","MIPT, Dolgoprudny, Russia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AIRI, Moscow, Russia","institution_ids":[]},{"raw_affiliation_string":"MIPT, Dolgoprudny, Russia","institution_ids":["https://openalex.org/I153845743"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5080806360","display_name":"Aleksandr I. Panov","orcid":"https://orcid.org/0000-0002-9747-3837"},"institutions":[{"id":"https://openalex.org/I153845743","display_name":"Moscow Institute of Physics and Technology","ror":"https://ror.org/00v0z9322","country_code":"RU","type":"education","lineage":["https://openalex.org/I153845743"]}],"countries":["RU"],"is_corresponding":false,"raw_author_name":"Aleksandr I. Panov","raw_affiliation_strings":["AIRI, Moscow, Russia","MIPT, Dolgoprudny, Russia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AIRI, Moscow, Russia","institution_ids":[]},{"raw_affiliation_string":"MIPT, Dolgoprudny, Russia","institution_ids":["https://openalex.org/I153845743"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I153845743"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.57608482,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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/T11367","display_name":"Particle accelerators and beam dynamics","score":0.9610999822616577,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace 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/T11367","display_name":"Particle accelerators and beam dynamics","score":0.9610999822616577,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace 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/T11005","display_name":"Radiation Effects in Electronics","score":0.9401999711990356,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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/T10363","display_name":"Low-power high-performance VLSI design","score":0.9269999861717224,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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/transformer","display_name":"Transformer","score":0.800000011920929},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7597000002861023},{"id":"https://openalex.org/keywords/robotics","display_name":"Robotics","score":0.45419999957084656},{"id":"https://openalex.org/keywords/online-learning","display_name":"Online learning","score":0.3903000056743622},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.3407999873161316}],"concepts":[{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.800000011920929},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7597000002861023},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6690000295639038},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5382000207901001},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.45419999957084656},{"id":"https://openalex.org/C2986087404","wikidata":"https://www.wikidata.org/wiki/Q15946010","display_name":"Online learning","level":2,"score":0.3903000056743622},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.3407999873161316},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33730000257492065},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.33219999074935913},{"id":"https://openalex.org/C133731056","wikidata":"https://www.wikidata.org/wiki/Q4917288","display_name":"Control engineering","level":1,"score":0.27950000762939453},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.2696000039577484},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.26350000500679016},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.25529998540878296},{"id":"https://openalex.org/C17098449","wikidata":"https://www.wikidata.org/wiki/Q176814","display_name":"Partially observable Markov decision process","level":4,"score":0.2529999911785126}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.3233/faia251080","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia251080","pdf_url":null,"source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"}],"best_oa_location":{"id":"doi:10.3233/faia251080","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia251080","pdf_url":null,"source":{"id":"https://openalex.org/S4210201731","display_name":"Frontiers in artificial intelligence and applications","issn_l":"0922-6389","issn":["0922-6389","1879-8314"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Frontiers in Artificial Intelligence and Applications","raw_type":"book-chapter"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0,72],"appearance":[1],"of":[2,13,26,37,53,97,130,178,193],"transformer-based":[3,47],"models":[4,48],"in":[5,15,32,114,128,139,162,203],"Reinforcement":[6],"Learning":[7],"(RL)":[8],"has":[9,20],"expanded":[10],"the":[11,38,51,54,65,82,88,94,98,102,105,112,126,198,212],"horizons":[12],"possibilities":[14],"robotics":[16],"tasks,":[17],"but":[18,180],"it":[19,135,196],"simultaneously":[21,86],"brought":[22],"a":[23,61,69,74,115,120,140,191],"wide":[24],"range":[25],"challenges":[27],"during":[28,93],"its":[29,131],"implementation,":[30],"especially":[31],"model-free":[33,123],"online":[34,117,138],"RL.":[35],"Some":[36],"existing":[39],"learning":[40],"algorithms":[41],"cannot":[42],"be":[43],"easily":[44],"implemented":[45],"with":[46,81,111],"due":[49],"to":[50,109,136,190,206],"instability":[52],"latter.":[55],"In":[56,101],"this":[57,122],"paper,":[58],"we":[59,167],"propose":[60],"method":[62],"that":[63,169],"uses":[64],"Accelerator":[66],"policy":[67],"as":[68,156,158],"transformer\u2019s":[70],"trainer.":[71],"Accelerator,":[73],"simpler":[75],"and":[76,133,143,152,164],"more":[77,141],"stable":[78,142,176],"model,":[79],"interacts":[80],"environment":[83,113],"independently":[84],"while":[85],"training":[87,177,183],"transformer":[89,107,127],"through":[90],"behavior":[91],"cloning":[92],"first":[95],"stage":[96],"proposed":[99],"algorithm.":[100],"second":[103],"stage,":[104],"pretrained":[106],"starts":[108],"interact":[110],"fully":[116],"setting.":[118],"As":[119],"result,":[121],"algorithm":[124,172],"accelerates":[125],"terms":[129],"performance":[132],"helps":[134],"train":[137],"faster":[144],"way.":[145],"By":[146],"conducting":[147],"experiments":[148],"on":[149,159,185],"both":[150],"state-based":[151],"image-based":[153,186],"ManiSkill":[154],"environments,":[155],"well":[157],"MuJoCo":[160],"tasks":[161],"MDP":[163],"POMDP":[165],"settings,":[166],"show":[168],"applying":[170],"our":[171],"not":[173],"only":[174],"enables":[175],"transformers":[179],"also":[181],"reduces":[182],"time":[184],"environments":[187],"by":[188],"up":[189],"factor":[192],"two.":[194],"Moreover,":[195],"decreases":[197],"required":[199],"replay":[200],"buffer":[201],"size":[202],"off-policy":[204],"methods":[205],"10\u201320":[207],"thousand,":[208],"which":[209],"significantly":[210],"lowers":[211],"overall":[213],"computational":[214],"demands.":[215]},"counts_by_year":[],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-24T00:00:00"}