{"id":"https://openalex.org/W4415428086","doi":"https://doi.org/10.3233/faia251203","title":"Improving Zero-Shot Generalization in Reinforcement Learning Through Abstract Representations","display_name":"Improving Zero-Shot Generalization in Reinforcement Learning Through Abstract Representations","publication_year":2025,"publication_date":"2025-10-21","ids":{"openalex":"https://openalex.org/W4415428086","doi":"https://doi.org/10.3233/faia251203"},"language":null,"primary_location":{"id":"doi:10.3233/faia251203","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia251203","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/faia251203","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Tsai Shuo Kuo","orcid":null},"institutions":[{"id":"https://openalex.org/I4400009020","display_name":"Institute of Science Tokyo","ror":"https://ror.org/05dqf9946","country_code":null,"type":"education","lineage":["https://openalex.org/I4400009020"]}],"countries":[],"is_corresponding":false,"raw_author_name":"Tsai Shuo Kuo","raw_affiliation_strings":["Institute of Science Tokyo"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Science Tokyo","institution_ids":["https://openalex.org/I4400009020"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5081854769","display_name":"Ryutaro Ichise","orcid":"https://orcid.org/0000-0001-8474-0150"},"institutions":[{"id":"https://openalex.org/I4400009020","display_name":"Institute of Science Tokyo","ror":"https://ror.org/05dqf9946","country_code":null,"type":"education","lineage":["https://openalex.org/I4400009020"]}],"countries":[],"is_corresponding":false,"raw_author_name":"Ryutaro Ichise","raw_affiliation_strings":["Institute of Science Tokyo"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Science Tokyo","institution_ids":["https://openalex.org/I4400009020"]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4400009020"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.33601193,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9538000226020813,"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":0.9538000226020813,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.7741000056266785},{"id":"https://openalex.org/keywords/overfitting","display_name":"Overfitting","score":0.707099974155426},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.6247000098228455},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5702999830245972},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.5443999767303467},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.49309998750686646},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.48910000920295715},{"id":"https://openalex.org/keywords/multi-task-learning","display_name":"Multi-task learning","score":0.34470000863075256},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.33889999985694885}],"concepts":[{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7741000056266785},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7378000020980835},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7278000116348267},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.707099974155426},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.6247000098228455},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6243000030517578},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5702999830245972},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5443999767303467},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.49309998750686646},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.48910000920295715},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.34470000863075256},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.33889999985694885},{"id":"https://openalex.org/C59656382","wikidata":"https://www.wikidata.org/wiki/Q191536","display_name":"Conjunction (astronomy)","level":2,"score":0.3149000108242035},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.31220000982284546},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.29910001158714294},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2874000072479248},{"id":"https://openalex.org/C116409475","wikidata":"https://www.wikidata.org/wiki/Q1385056","display_name":"External Data Representation","level":2,"score":0.2797999978065491},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2782000005245209},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.27559998631477356},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.26910001039505005},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.2678000032901764},{"id":"https://openalex.org/C2986563244","wikidata":"https://www.wikidata.org/wiki/Q6822310","display_name":"Learning to learn","level":2,"score":0.2669000029563904},{"id":"https://openalex.org/C24138899","wikidata":"https://www.wikidata.org/wiki/Q17141258","display_name":"Instance-based learning","level":3,"score":0.265500009059906},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.26179999113082886},{"id":"https://openalex.org/C2781002164","wikidata":"https://www.wikidata.org/wiki/Q6822311","display_name":"Meta learning (computer science)","level":3,"score":0.2605000138282776}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.3233/faia251203","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia251203","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/faia251203","is_oa":true,"landing_page_url":"https://doi.org/10.3233/faia251203","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":{"Reinforcement":[0],"learning":[1,28,38,65,76],"algorithms":[2],"suffer":[3],"from":[4],"zero-shot":[5,189],"generalization,":[6],"where":[7,151],"a":[8,86,142,146],"learned":[9,161],"policy":[10],"is":[11,22],"tested":[12],"in":[13,92,100,203,212],"new,":[14],"out-of-distribution":[15],"environments.":[16],"One":[17],"of":[18,25],"the":[19,23,26,74,93,101,106,110,115,160,165,169,179,186],"key":[20],"reasons":[21],"overfitting":[24],"representation":[27,75,143,162],"model.":[29],"Previous":[30],"studies":[31],"have":[32],"used":[33,188,202],"data":[34,218],"augmentation":[35],"or":[36,51,136],"contrastive":[37],"to":[39,59,78,89,163],"address":[40],"this":[41,120],"overfitting.":[42],"Yet":[43],"these":[44],"methods":[45,208],"often":[46],"rely":[47],"on":[48,119,185],"task-specific":[49],"augmentations":[50,135],"hand-picked":[52],"negative":[53,137],"samples,":[54],"which":[55],"are":[56],"not":[57],"robust":[58],"various":[60],"tasks.":[61],"We":[62],"hypothesize":[63],"that":[64,131],"more":[66],"abstract":[67,155],"representations":[68],"by":[69,182],"ignoring":[70,173],"pixel-level":[71,174],"details":[72],"allows":[73],"model":[77],"generalize":[79],"better.":[80],"Moreover,":[81],"when":[82],"an":[83,129,154],"agent":[84],"encounters":[85],"scene":[87,111],"similar":[88],"one":[90],"seen":[91],"training":[94],"environment":[95],"but":[96],"with":[97,125,145,205,214],"slight":[98],"variations":[99],"background,":[102],"it":[103,152],"will":[104],"ignore":[105],"differences":[107],"and":[108,209],"treat":[109],"as":[112,217],"invariant,":[113],"making":[114],"correct":[116],"decision.":[117],"Based":[118],"idea,":[121],"we":[122,140,158],"introduce":[123],"Dream":[124],"Abstractions":[126],"(Dr.":[127],"Abs),":[128],"algorithm":[130,181],"requires":[132],"no":[133],"custom":[134],"samples.":[138],"First,":[139],"train":[141],"network":[144],"joint":[147],"embedding":[148],"predictive":[149],"architecture,":[150],"learns":[153],"representation.":[156],"Next,":[157],"use":[159],"regularize":[164],"latent":[166],"states":[167],"inside":[168],"reinforcement-learning":[170],"agent.":[171],"By":[172],"details,":[175],"our":[176],"method":[177],"outperforms":[178],"baseline":[180],"approximately":[183],"22%":[184],"widely":[187],"generalization":[190,195],"setup,":[191],"thereby":[192],"enhancing":[193],"its":[194],"ability.":[196],"Furthermore,":[197],"Dr.":[198],"Abs":[199],"can":[200,210],"be":[201],"conjunction":[204],"other":[206],"RL":[207],"work":[211],"tandem":[213],"techniques":[215],"such":[216],"augmentation.":[219]},"counts_by_year":[],"updated_date":"2026-07-25T17:18:41.151588","created_date":"2025-10-24T00:00:00"}