{"id":"W4380136285","doi":"10.48550/arxiv.2306.04923","title":"Unconstrained Online Learning with Unbounded Losses","year":2023,"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":"","funders":"Natural Sciences and Engineering Research Council of Canada; National Science Foundation","keywords":"Regret; Lipschitz continuity; Bounded function; Saddle point; Mathematics; Leverage (statistics); Domain (mathematical analysis); Combinatorics; Upper and lower bounds; Matching (statistics); Saddle; Mathematical optimization; Discrete mathematics; Computer science; Pure mathematics; Mathematical analysis","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001531814,0.0004804212,0.0007393027,0.001075921,0.0004165256,0.0004011045,0.002395417,0.0003614036,0.0004667524],"category_scores_gemma":[0.002691114,0.0003958196,0.000263222,0.002864551,0.0008399286,0.0004133118,0.002285095,0.001876451,0.001043846],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003327557,"about_ca_system_score_gemma":0.0006302221,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002172439,"about_ca_topic_score_gemma":0.000825802,"domain_scores_codex":[0.995338,0.0005423452,0.0004688002,0.002041574,0.0008904367,0.00071886],"domain_scores_gemma":[0.9941815,0.00254373,0.0005160239,0.001486937,0.0009263876,0.0003453857],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000280553,0.00007994001,0.01613061,0.00003155522,0.0001395974,0.001946148,0.0001368966,0.976816,0.00001178027,0.001272077,0.0004685183,0.002686354],"study_design_scores_gemma":[0.00384405,0.0008689719,0.01206165,0.0004291502,0.0002126419,0.00004720874,0.008022157,0.6748186,0.0002704197,0.2883584,0.009038627,0.002028157],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6312563,0.00003790617,0.3653558,0.0002995143,0.000499083,0.0005215043,0.0001555852,0.000543246,0.001331024],"genre_scores_gemma":[0.941581,0.0001206149,0.0009507027,0.00001997134,0.0001717785,0.000001467265,0.0001013359,0.00006288779,0.05699021],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3644051,"threshold_uncertainty_score":0.9998494,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3545191683132599,"score_gpt":0.3271747835432784,"score_spread":0.02734438476998141,"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."}}