{"id":"https://openalex.org/W7160940016","doi":"https://doi.org/10.48550/arxiv.2605.10137","title":"PFN-TS: Thompson Sampling for Contextual Bandits via Prior-Data Fitted Networks","display_name":"PFN-TS: Thompson Sampling for Contextual Bandits via Prior-Data Fitted Networks","publication_year":2026,"publication_date":"2026-05-11","ids":{"openalex":"https://openalex.org/W7160940016","doi":"https://doi.org/10.48550/arxiv.2605.10137"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.10137","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.10137","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.10137","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135915655","display_name":"Yan Shuo Tan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tan, Yan Shuo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135978753","display_name":"Kenyon Ng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ng, Kenyon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125491547","display_name":"Ruizhe Deng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Deng, Ruizhe","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135958145","display_name":"Sumetha Loganathan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Loganathan, Sumetha","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135924015","display_name":"Qiong Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Qiong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5036008511","display_name":"Bibhas Chakraborty","orcid":"https://orcid.org/0000-0002-7366-0478"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chakraborty, Bibhas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":[],"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/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.9714999794960022,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.9714999794960022,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.008899999782443047,"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":0.0027000000700354576,"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/thompson-sampling","display_name":"Thompson sampling","score":0.7455000281333923},{"id":"https://openalex.org/keywords/regret","display_name":"Regret","score":0.7404000163078308},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.5411999821662903},{"id":"https://openalex.org/keywords/estimator","display_name":"Estimator","score":0.5386999845504761},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.5192999839782715},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.47679999470710754},{"id":"https://openalex.org/keywords/posterior-probability","display_name":"Posterior probability","score":0.46369999647140503},{"id":"https://openalex.org/keywords/importance-sampling","display_name":"Importance sampling","score":0.45170000195503235},{"id":"https://openalex.org/keywords/limit","display_name":"Limit (mathematics)","score":0.3840000033378601}],"concepts":[{"id":"https://openalex.org/C73602740","wikidata":"https://www.wikidata.org/wiki/Q7795822","display_name":"Thompson sampling","level":3,"score":0.7455000281333923},{"id":"https://openalex.org/C50817715","wikidata":"https://www.wikidata.org/wiki/Q79895177","display_name":"Regret","level":2,"score":0.7404000163078308},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5813000202178955},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.5411999821662903},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.5386999845504761},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.5192999839782715},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.47679999470710754},{"id":"https://openalex.org/C57830394","wikidata":"https://www.wikidata.org/wiki/Q278079","display_name":"Posterior probability","level":3,"score":0.46369999647140503},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.45170000195503235},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.450300008058548},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.41269999742507935},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.3840000033378601},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.3725999891757965},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.366100013256073},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33239999413490295},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.3199999928474426},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.3192000091075897},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.31869998574256897},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.30720001459121704},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.3066999912261963},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.29919999837875366},{"id":"https://openalex.org/C187691185","wikidata":"https://www.wikidata.org/wiki/Q2020720","display_name":"Grid","level":2,"score":0.2978000044822693},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.2888000011444092},{"id":"https://openalex.org/C68022304","wikidata":"https://www.wikidata.org/wiki/Q842217","display_name":"Bayes estimator","level":3,"score":0.2800000011920929},{"id":"https://openalex.org/C149569020","wikidata":"https://www.wikidata.org/wiki/Q25098598","display_name":"Bayesian average","level":5,"score":0.2797999978065491},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.26030001044273376},{"id":"https://openalex.org/C2776291640","wikidata":"https://www.wikidata.org/wiki/Q2912517","display_name":"Value (mathematics)","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.10137","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.10137","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.10137","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.10137","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Thompson":[0,64,78],"sampling":[1,65,79],"is":[2,192],"a":[3,15,19,53,77,90,102,139],"widely":[4],"used":[5,117],"strategy":[6],"for":[7,42],"contextual":[8],"bandits:":[9],"at":[10,194],"each":[11],"round,":[12],"it":[13],"samples":[14,88],"reward":[16,72],"function":[17],"from":[18,101],"Bayesian":[20,48,140],"posterior":[21,49,84,99],"and":[22,36,122,137,168,176,179],"acts":[23],"greedily":[24],"under":[25,151],"that":[26,81,143],"sample.":[27],"Prior-data":[28],"fitted":[29],"networks":[30],"(PFNs),":[31],"such":[32],"as":[33],"TabPFN":[34],"v2+":[35],"TabICL":[37],"v2,":[38],"are":[39],"attractive":[40],"candidates":[41],"this":[43],"purpose":[44],"because":[45],"they":[46],"approximate":[47],"predictive":[50,92,115],"distributions":[51],"in":[52,118,186],"single":[54],"forward":[55],"pass.":[56],"However,":[57],"PFNs":[58],"predict":[59],"noisy":[60],"future":[61],"rewards,":[62,178],"while":[63],"requires":[66],"uncertainty":[67],"over":[68],"the":[69,112,133,152,161,181],"latent":[70],"mean":[71],"function.":[73],"We":[74,129],"propose":[75],"PFN-TS,":[76],"algorithm":[80],"converts":[82],"PFN":[83,153],"predictives":[85],"into":[86,147],"mean-reward":[87],"using":[89],"subsampled":[91,134],"central":[93],"limit":[94],"theorem.":[95],"The":[96],"method":[97],"estimates":[98],"variance":[100,135],"geometric":[103],"grid":[104],"of":[105,132],"$O(\\log":[106],"n)$":[107],"dataset":[108],"prefixes":[109],"rather":[110],"than":[111],"full":[113],"$O(n)$":[114],"sequence":[116],"previous":[119],"predictive-sequence":[120],"approaches,":[121],"reuses":[123],"TabICL's":[124],"cached":[125],"representations":[126],"across":[127,165],"rounds.":[128],"prove":[130],"consistency":[131],"estimator":[136],"give":[138],"regret":[141,146,150],"bound":[142],"decomposes":[144],"PFN-TS":[145,159],"exact":[148],"posterior-sampling":[149],"prior":[154],"plus":[155],"approximation":[156],"terms.":[157],"Empirically,":[158],"achieves":[160],"best":[162],"average":[163],"rank":[164],"nonlinear":[166],"synthetic":[167],"OpenML":[169],"classification-to-bandit":[170],"benchmarks,":[171],"remains":[172],"competitive":[173],"on":[174],"linear":[175],"BART-generated":[177],"attains":[180],"highest":[182],"estimated":[183],"policy":[184],"value":[185],"an":[187],"offline":[188],"mobile-health":[189],"evaluation.":[190],"Code":[191],"available":[193],"https://anonymous.4open.science/r/PFN_TS-36ED/.":[195]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-13T00:00:00"}