{"id":"https://openalex.org/W4417354635","doi":"https://doi.org/10.48550/arxiv.2506.10629","title":"Task Adaptation from Skills: Information Geometry, Disentanglement, and New Objectives for Unsupervised Reinforcement Learning","display_name":"Task Adaptation from Skills: Information Geometry, Disentanglement, and New Objectives for Unsupervised Reinforcement Learning","publication_year":2025,"publication_date":"2025-06-12","ids":{"openalex":"https://openalex.org/W4417354635","doi":"https://doi.org/10.48550/arxiv.2506.10629"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2506.10629","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.10629","pdf_url":"https://arxiv.org/pdf/2506.10629","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","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2506.10629","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101482568","display_name":"Yucheng Yang","orcid":"https://orcid.org/0009-0002-6328-6360","observed_orcids":["https://orcid.org/0009-0002-6328-6360"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Yucheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039076312","display_name":"Tianyi Zhou","orcid":"https://orcid.org/0000-0001-5348-0632","observed_orcids":["https://orcid.org/0000-0001-5348-0632"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Tianyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019915847","display_name":"Qiang He","orcid":"https://orcid.org/0000-0002-1820-6141","observed_orcids":["https://orcid.org/0000-0002-1820-6141"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Qiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022456493","display_name":"Lei Han","orcid":"https://orcid.org/0000-0002-7389-782X","observed_orcids":["https://orcid.org/0000-0002-7389-782X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Lei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022601535","display_name":"Mykola Pechenizkiy","orcid":"https://orcid.org/0000-0003-4955-0743","observed_orcids":["https://orcid.org/0000-0003-4955-0743"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pechenizkiy, Mykola","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100400497","display_name":"Meng Fang","orcid":"https://orcid.org/0000-0001-6745-286X","observed_orcids":["https://orcid.org/0000-0001-6745-286X","https://orcid.org/0000-0003-0793-9187"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Meng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":2,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0003000000142492354,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.0,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/unsupervised-reinforcement-learning","display_name":"unsupervised reinforcement learning","score":0.996999979019165},{"id":"https://openalex.org/keywords/information-geometry","display_name":"information geometry","score":0.902999997138977},{"id":"https://openalex.org/keywords/wasserstein-distance","display_name":"Wasserstein distance","score":0.9509999752044678}],"concepts":[{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.7473000288009644},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.7024999856948853},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6837999820709229},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.680899977684021},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6330000162124634},{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.5705999732017517},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.5432999730110168},{"id":"https://openalex.org/C112313634","wikidata":"https://www.wikidata.org/wiki/Q7886648","display_name":"Complement (music)","level":5,"score":0.5406000018119812},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5159000158309937},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.48840001225471497},{"id":"https://openalex.org/C171752962","wikidata":"https://www.wikidata.org/wiki/Q255166","display_name":"Kullback\u2013Leibler divergence","level":2,"score":0.43709999322891235},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40950000286102295},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.38190001249313354},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.3605000078678131},{"id":"https://openalex.org/C52622258","wikidata":"https://www.wikidata.org/wiki/Q131222","display_name":"Information theory","level":2,"score":0.31450000405311584},{"id":"https://openalex.org/C113336015","wikidata":"https://www.wikidata.org/wiki/Q574010","display_name":"Complete information","level":2,"score":0.2800000011920929},{"id":"https://openalex.org/C8038995","wikidata":"https://www.wikidata.org/wiki/Q1152135","display_name":"Unsupervised learning","level":2,"score":0.2712000012397766},{"id":"https://openalex.org/C2779136372","wikidata":"https://www.wikidata.org/wiki/Q10283002","display_name":"Information flow","level":2,"score":0.25940001010894775},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.2590999901294708},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2563999891281128},{"id":"https://openalex.org/C2780898871","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Performance metric","level":2,"score":0.2508000135421753}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2506.10629","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.10629","pdf_url":"https://arxiv.org/pdf/2506.10629","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.2506.10629","doi":"https://doi.org/10.48550/arxiv.2506.10629","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.10629","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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2506.10629","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.10629","pdf_url":"https://arxiv.org/pdf/2506.10629","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":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Unsupervised":[0],"reinforcement":[1],"learning":[2],"(URL)":[3],"aims":[4],"to":[5,65,125,129,139,142],"learn":[6],"general":[7],"skills":[8,28,39,61],"for":[9,155],"unseen":[10],"downstream":[11,43,66,96,143,156],"tasks.":[12],"Mutual":[13],"Information":[14],"Skill":[15],"Learning":[16],"(MISL)":[17],"addresses":[18],"URL":[19],"by":[20],"maximizing":[21],"the":[22,55,111,122],"mutual":[23],"information":[24,115],"between":[25,93],"states":[26],"and":[27,57,95,146],"but":[29,69],"lacks":[30],"sufficient":[31],"theoretical":[32,48],"analysis,":[33],"e.g.,":[34],"how":[35],"well":[36],"its":[37],"learned":[38,60],"can":[40,169],"initialize":[41],"a":[42,82,106,130],"task's":[44],"policy.":[45],"Our":[46],"new":[47,107],"analysis":[49,124],"in":[50,114],"this":[51],"paper":[52],"shows":[53],"that":[54,109,168],"diversity":[56],"separability":[58],"of":[59,150],"are":[62],"fundamentally":[63],"critical":[64],"task":[67,97,144],"adaptation":[68,98,145],"MISL":[70],"does":[71],"not":[72],"necessarily":[73],"guarantee":[74],"these":[75],"properties.":[76],"To":[77],"complement":[78],"MISL,":[79],"we":[80,88,104],"propose":[81,162],"novel":[83,131],"disentanglement":[84],"metric":[85],"LSEPIN.":[86],"Moreover,":[87],"build":[89],"an":[90],"information-geometric":[91],"connection":[92],"LSEPIN":[94],"cost.":[99],"For":[100],"better":[101],"geometric":[102,123],"properties,":[103],"investigate":[105],"strategy":[108],"replaces":[110],"KL":[112],"divergence":[113],"geometry":[116],"with":[117],"Wasserstein":[118,164],"distance.":[119],"We":[120,160],"extend":[121],"it,":[126],"which":[127],"leads":[128],"skill-learning":[132],"objective":[133],"WSEP.":[134],"It":[135],"is":[136,148],"theoretically":[137,170],"justified":[138],"be":[140],"helpful":[141],"it":[147],"capable":[149],"discovering":[151],"more":[152],"initial":[153,174],"policies":[154],"tasks":[157],"than":[158],"MISL.":[159],"finally":[161],"another":[163],"distance-based":[165],"algorithm":[166],"PWSEP":[167],"discover":[171],"all":[172],"optimal":[173],"policies.":[175]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2025-10-10T00:00:00"}