{"id":"https://openalex.org/W7220432290","doi":"https://doi.org/10.48550/arxiv.2610.03667","title":"Planning to Learn","display_name":"Planning to Learn","publication_year":2026,"publication_date":"2026-10-02","ids":{"openalex":"https://openalex.org/W7220432290","doi":"https://doi.org/10.48550/arxiv.2610.03667"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.03667","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.03667","pdf_url":"https://arxiv.org/pdf/2610.03667","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","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.03667","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5015899120","display_name":"Ian Osband","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Osband, Ian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5015899120"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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":0.7778000235557556,"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":0.7778000235557556,"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":0.12999999523162842,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.03460000082850456,"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/policy-gradient","display_name":"policy gradient","score":0.6610000133514404},{"id":"https://openalex.org/keywords/cross-entropy-loss","display_name":"cross-entropy loss","score":0.906000018119812},{"id":"https://openalex.org/keywords/learning-rate-schedules","display_name":"learning rate scheduling","score":0.3610000014305115},{"id":"https://openalex.org/keywords/noisy-labels","display_name":"label noise","score":0.8709999918937683},{"id":"https://openalex.org/keywords/training-dynamics","display_name":"training dynamics","score":0.36399999260902405},{"id":"https://openalex.org/keywords/myopia","display_name":"myopia","score":0.3919999897480011},{"id":"https://openalex.org/keywords/llm-post-training","display_name":"LLM post-training","score":0.5910000205039978},{"id":"https://openalex.org/keywords/image-classification","display_name":"image classification","score":0.3109999895095825}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7188000082969666},{"id":"https://openalex.org/C28761237","wikidata":"https://www.wikidata.org/wiki/Q7805321","display_name":"Time horizon","level":2,"score":0.6353999972343445},{"id":"https://openalex.org/C190502265","wikidata":"https://www.wikidata.org/wiki/Q17069496","display_name":"MNIST database","level":3,"score":0.49070000648498535},{"id":"https://openalex.org/C159176650","wikidata":"https://www.wikidata.org/wiki/Q43261","display_name":"Horizon","level":2,"score":0.4844000041484833},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4722999930381775},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.44119998812675476},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3993000090122223},{"id":"https://openalex.org/C115680565","wikidata":"https://www.wikidata.org/wiki/Q5977448","display_name":"Gradient method","level":2,"score":0.3693999946117401},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3449999988079071},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.33809998631477356},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.3285999894142151},{"id":"https://openalex.org/C520416788","wikidata":"https://www.wikidata.org/wiki/Q5419229","display_name":"Exact solutions in general relativity","level":2,"score":0.30660000443458557},{"id":"https://openalex.org/C188116033","wikidata":"https://www.wikidata.org/wiki/Q2664563","display_name":"Q-learning","level":3,"score":0.2827000021934509},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.2822999954223633},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.27810001373291016},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26499998569488525}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.03667","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.03667","pdf_url":"https://arxiv.org/pdf/2610.03667","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.2610.03667","doi":"https://doi.org/10.48550/arxiv.2610.03667","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.03667","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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.03667","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.03667","pdf_url":"https://arxiv.org/pdf/2610.03667","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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7220432290.pdf","grobid_xml":"https://content.openalex.org/works/W7220432290.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W2799151646","https://openalex.org/W4385462263","https://openalex.org/W4391631327","https://openalex.org/W4414281281","https://openalex.org/W7127577415","https://openalex.org/W7138216039","https://openalex.org/W7140172760","https://openalex.org/W7140198410","https://openalex.org/W7196953391"],"related_works":["https://openalex.org/W7131079924","https://openalex.org/W4226317014","https://openalex.org/W7169864415","https://openalex.org/W7129438424","https://openalex.org/W7134258915","https://openalex.org/W7223053544","https://openalex.org/W7163638543","https://openalex.org/W7167239199","https://openalex.org/W2942218726","https://openalex.org/W7164914334"],"abstract_inverted_index":{"Policy-gradient":[0],"methods":[1],"are":[2,17],"central":[3],"to":[4,45,67],"modern":[5],"reinforcement":[6],"learning,":[7],"including":[8],"LLM":[9],"post-training.":[10],"When":[11],"they":[12],"struggle,":[13],"the":[14,41,46,55,94,115,135,142,147,173,189,203],"usual":[15],"suspects":[16],"exploration,":[18],"credit":[19],"assignment":[20],"and":[21,49,60,181,187,202],"action-sampling":[22],"noise.":[23,208],"Classification":[24],"has":[25],"none":[26],"of":[27],"them.":[28],"A":[29],"classifier":[30],"is":[31,40,53,58,76,112,134],"a":[32,150,166,198],"policy":[33,56,64,132,159],"whose":[34],"expected":[35,71],"reward,":[36],"its":[37,123],"\\emph{expected":[38],"accuracy},":[39],"probability":[42],"it":[43,78,85,170],"assigns":[44],"correct":[47],"label,":[48],"because":[50],"that":[51,144,153,175],"label":[52,207],"known,":[54],"gradient":[57,65,75,133,160],"exact":[59,63,74,131,158],"smooth.":[61],"Yet":[62],"loses":[66],"cross-entropy,":[68],"even":[69],"on":[70,103,182],"accuracy.":[72],"The":[73],"myopic:":[77],"values":[79],"an":[80,99,118],"update":[81,90],"only":[82],"by":[83],"what":[84],"buys":[86],"now,":[87],"but":[88],"each":[89,177],"also":[91],"sets":[92],"where":[93],"next":[95],"one":[96],"starts,":[97],"so":[98],"update's":[100],"value":[101],"depends":[102],"how":[104],"much":[105],"learning":[106,143,200],"remains.":[107],"Viewed":[108],"this":[109,139],"way,":[110],"cross-entropy":[111,156,196],"patient":[113],"accuracy,":[114],"total":[116,140],"error":[117],"example":[119],"would":[120],"pay":[121],"if":[122],"log-odds":[124],"rose":[125],"at":[126,141,197],"unit":[127],"speed":[128],"forever,":[129],"while":[130],"zero-horizon":[136],"limit.":[137],"Truncating":[138],"remains":[145],"yields":[146],"horizon":[148,190],"loss,":[149],"one-line":[151],"change":[152],"moves":[154],"from":[155],"toward":[157],"as":[161],"training":[162],"runs":[163],"out.":[164],"In":[165],"simple":[167],"allocation":[168],"model,":[169],"provably":[171],"escapes":[172],"trap":[174],"catches":[176],"endpoint.":[178],"On":[179],"MNIST":[180],"ImageNet":[183],"with":[184,206],"ResNet-50,":[185],"ResNet-101":[186],"ViT-S/16,":[188],"loss":[191],"improves":[192],"top-1":[193],"accuracy":[194],"over":[195],"flat":[199],"rate,":[201],"gain":[204],"grows":[205]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-06T00:00:00"}