{"id":"W2552858844","doi":"10.48550/arxiv.1702.03040","title":"Following the Leader and Fast Rates in Linear Prediction: Curved Constraint Sets and Other Regularities","year":2017,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; Alberta Innovates; University of Alberta","keywords":"Sublinear function; Regret; Bounded function; Logarithm; Boundary (topology); Domain (mathematical analysis); Constraint (computer-aided design); Regular polygon; Curvature; Computer science; Algorithm; Mathematical optimization; Mathematics; Upper and lower bounds; Discrete mathematics; Geometry; Mathematical analysis","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001187052,0.0001105562,0.0001790258,0.0001452647,0.0006613257,0.0002868297,0.0005354048,0.00007147066,0.00005534586],"category_scores_gemma":[0.0007860762,0.00008076173,0.00004940952,0.0002121232,0.001000242,0.0006749554,0.0003120898,0.000211333,0.00001771622],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000290533,"about_ca_system_score_gemma":0.00004225774,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001253413,"about_ca_topic_score_gemma":0.0003203874,"domain_scores_codex":[0.9987867,0.0001588555,0.0001568849,0.0004587775,0.0002046475,0.0002341319],"domain_scores_gemma":[0.9985754,0.0005824249,0.0000998758,0.0005683845,0.0000878324,0.00008605397],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001426151,0.00005978498,0.9687643,0.0000133709,0.00008422316,0.0004894467,0.002214529,0.006344267,0.0001641741,0.009706487,0.0004511295,0.01156567],"study_design_scores_gemma":[0.004165654,0.0001650699,0.4408942,0.0001301768,0.00003968725,0.00004995877,0.02702905,0.4172578,0.0005375557,0.1019753,0.007242475,0.0005131307],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9859969,0.0001296246,0.01155069,0.0006097321,0.0001751861,0.0001973417,0.00002508489,0.00002168723,0.001293728],"genre_scores_gemma":[0.9954814,0.00004867704,0.0001362678,0.00004622564,0.00003739118,5.509544e-7,5.437981e-7,0.000007330536,0.004241559],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5278702,"threshold_uncertainty_score":0.5086449,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2529146763771616,"score_gpt":0.3257650853383418,"score_spread":0.07285040896118011,"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."}}