{"id":"W7117323897","doi":"10.1007/s10994-025-06943-6","title":"Extended UCB Policies for Multi-armed Bandit Problems","year":2025,"lang":"en","type":"article","venue":"Machine Learning","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Novelis (Canada)","funders":"","keywords":"Regret; Process (computing); Markov decision process; Simplicity; Order (exchange); Reinforcement learning; Extension (predicate logic)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.002574199,0.0002084534,0.0003847441,0.0006982075,0.0006478426,0.0003525226,0.0008694151,0.00009539142,0.0003537582],"category_scores_gemma":[0.01145622,0.0001551649,0.0001704727,0.001243871,0.0001741811,0.0002626498,0.0003560881,0.0004748387,0.0001396621],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008663331,"about_ca_system_score_gemma":0.0001056955,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001419023,"about_ca_topic_score_gemma":0.000160431,"domain_scores_codex":[0.9969368,0.0002802921,0.0005786738,0.0006393566,0.0009398281,0.0006250275],"domain_scores_gemma":[0.9967943,0.002039803,0.0001791714,0.000470259,0.0004006195,0.000115799],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003365572,0.0004255272,0.0610071,0.00013722,0.0001719397,0.00002295832,0.003197398,0.1258164,0.00570742,0.003856928,0.01229052,0.7870301],"study_design_scores_gemma":[0.002055271,0.0001615,0.01831249,0.00004244594,0.00001160622,0.000004483434,0.0005348827,0.5994303,0.001268886,0.01722389,0.3606986,0.0002556375],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00972465,0.000982565,0.9819115,0.002964228,0.0004466046,0.0009338223,0.00002990021,0.0002336864,0.002773053],"genre_scores_gemma":[0.7947925,0.00005337385,0.03199154,0.000443744,0.0002237557,0.0002994422,0.00004204069,0.00005370644,0.1720999],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9499199,"threshold_uncertainty_score":0.9968707,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1527085351058629,"score_gpt":0.4869636691665585,"score_spread":0.3342551340606956,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}