{"id":"W3210467483","doi":"10.48550/arxiv.2111.00870","title":"Statistical Consequences of Dueling Bandits","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Regret; Computer science; Thompson sampling; Operations research; Management science; Machine learning; Mathematics; 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":[0.09571703,0.001453014,0.003406081,0.001532888,0.001782968,0.004111645,0.002827137,0.004078382,0.004739764],"category_scores_gemma":[0.3465251,0.001164246,0.001587638,0.001867723,0.009445325,0.006273359,0.00340162,0.006596306,0.0005980058],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002607626,"about_ca_system_score_gemma":0.002035023,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003711591,"about_ca_topic_score_gemma":0.002086501,"domain_scores_codex":[0.9425356,0.04487558,0.002261163,0.005083817,0.004162192,0.001081696],"domain_scores_gemma":[0.5952065,0.360305,0.01764315,0.01900478,0.006557437,0.001283105],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0006990305,0.0001765271,0.009797426,0.0003438425,0.0004388511,0.0003895562,0.0006483748,0.4282819,0.0007610755,0.5190964,0.003238122,0.0361289],"study_design_scores_gemma":[0.000167531,0.0001974775,0.002270953,0.0001246358,0.00006694471,0.0000838732,0.0001535463,0.5367886,0.0006597192,0.4581442,0.001281703,0.00006078528],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0843965,0.001355258,0.901706,0.004821851,0.0002810404,0.0004214874,0.0005131982,0.0003936157,0.006111002],"genre_scores_gemma":[0.8588668,0.0007577466,0.1324482,0.002012928,0.0002710818,0.001152809,0.0004356987,0.0001974735,0.00385714],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.09571703,"threshold_uncertainty_score":0.506206,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3207690795247978,"score_gpt":0.3360774280993675,"score_spread":0.01530834857456964,"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."}}