{"id":"W2142971854","doi":"10.1016/j.tcs.2009.01.016","title":"Exploration–exploitation tradeoff using variance estimates in multi-armed bandits","year":2009,"lang":"en","type":"article","venue":"Theoretical Computer Science","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":566,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Regret; Variance (accounting); Stochastic game; Upper and lower bounds; Logarithm; Mathematical optimization; Computer science; Multi-armed bandit; Interval (graph theory); Mathematics; Mathematical economics; Machine learning; Economics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01741206,0.001801049,0.003292357,0.001939547,0.001008865,0.004078803,0.00246353,0.004149533,0.002009197],"category_scores_gemma":[0.08031216,0.001942511,0.001007089,0.001705955,0.002668021,0.006750336,0.003856997,0.003234298,0.0004377603],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001378395,"about_ca_system_score_gemma":0.001663079,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001100996,"about_ca_topic_score_gemma":0.0008045105,"domain_scores_codex":[0.9900691,0.007260599,0.0004701544,0.0006379086,0.001006346,0.0005558263],"domain_scores_gemma":[0.9136565,0.07833941,0.002657933,0.00258174,0.002190966,0.0005735562],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0005399436,0.00009468298,0.00171695,0.0002333121,0.0002439849,0.00007355642,0.0001621639,0.8388136,0.001278294,0.111198,0.0009046881,0.04474088],"study_design_scores_gemma":[0.0000448597,0.0000698111,0.0002316226,0.00003990339,0.00002981559,0.00003173484,0.00001833976,0.9476445,0.0004835272,0.05120246,0.0001807996,0.00002271905],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03525495,0.001214849,0.959457,0.0008320826,0.00005525736,0.0000427771,0.00005216048,0.000178787,0.002912093],"genre_scores_gemma":[0.8497404,0.0009737169,0.1452886,0.000284409,0.0002174992,0.0003045098,0.0001220478,0.0002014138,0.002867379],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01741206,"threshold_uncertainty_score":0.09208488,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2025090972907981,"score_gpt":0.4631392662511002,"score_spread":0.2606301689603022,"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."}}