{"id":"W3118438270","doi":"10.1214/23-aos2315","title":"Relaxing the i.i.d. assumption: Adaptively minimax optimal regret via root-entropic regularization","year":2023,"lang":"en","type":"article","venue":"The Annals of Statistics","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; University of Toronto","funders":"","keywords":"Regret; Minimax; Mathematics; Mathematical optimization; Regularization (linguistics); Entropy (arrow of time); Adversarial system; Computer science; Artificial intelligence; Statistics","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":[],"consensus_categories":[],"category_scores_codex":[0.00399619,0.0001507093,0.0002631421,0.0002094887,0.0004780352,0.0001670131,0.001205576,0.00006638638,0.0002424743],"category_scores_gemma":[0.007074946,0.00008143314,0.00007952615,0.001634909,0.0005279714,0.0002612623,0.0003434772,0.0002871674,0.0005294697],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000221613,"about_ca_system_score_gemma":0.00009412037,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001622301,"about_ca_topic_score_gemma":0.00002069151,"domain_scores_codex":[0.9957457,0.0005067955,0.0006784822,0.0003440525,0.002297171,0.0004278486],"domain_scores_gemma":[0.9926569,0.004838029,0.0004380071,0.0008413785,0.001145139,0.0000804996],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0005077708,0.0001244015,0.001478806,0.00003899744,0.000213215,0.000113814,0.005397922,0.09701305,0.001443783,0.06985162,0.4522492,0.3715675],"study_design_scores_gemma":[0.0005076634,0.0003981547,0.09779119,0.00005273684,0.00003157512,0.00002061242,0.002496164,0.4056778,0.002779147,0.4565051,0.03343838,0.0003014227],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01007596,0.0001982746,0.9800175,0.008340136,0.0002500227,0.0003420785,0.0003429041,0.00005401528,0.0003790429],"genre_scores_gemma":[0.924117,0.0008707478,0.04112681,0.000422012,0.0006813922,0.00006743817,0.000146734,0.00008510833,0.03248273],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9388908,"threshold_uncertainty_score":0.8469883,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3214253860286645,"score_gpt":0.4722292110110216,"score_spread":0.1508038249823571,"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."}}