{"id":"W2146807381","doi":"","title":"PAC-Bayesian Analysis of Contextual Bandits","year":2011,"lang":"la","type":"article","venue":"MPG.PuRe (Max Planck Society)","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":40,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"","keywords":"Regret; Logarithm; Upper and lower bounds; Scaling; Bayesian probability; Computer science; Task (project management); Combinatorics; Mathematics; Algorithm; Artificial intelligence; Machine learning","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003324958,0.0008113331,0.002324453,0.001062705,0.0005797992,0.0002751302,0.002668455,0.0009129,0.01404912],"category_scores_gemma":[0.001475158,0.0006783378,0.002388561,0.009074482,0.001433234,0.0008480049,0.0006548336,0.001223682,0.0007410471],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002475991,"about_ca_system_score_gemma":0.0005755675,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003493142,"about_ca_topic_score_gemma":0.0005564822,"domain_scores_codex":[0.9895065,0.0006123513,0.002102149,0.001836334,0.004257293,0.00168535],"domain_scores_gemma":[0.9921967,0.00240957,0.001190624,0.002121509,0.001320883,0.0007607531],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.002346171,0.004867786,0.2433715,0.0005786284,0.03880424,0.001063153,0.1579727,0.004257509,0.003402789,0.007171271,0.3677955,0.1683687],"study_design_scores_gemma":[0.01514826,0.004632344,0.3654774,0.0007835436,0.01308016,0.000105246,0.1002825,0.2416027,0.01485556,0.03179546,0.2043531,0.00788369],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3133975,0.01532956,0.5645192,0.002086153,0.008075756,0.004816557,0.01110442,0.0007230752,0.07994782],"genre_scores_gemma":[0.9787253,0.0005707167,0.006141969,0.0003293398,0.0005233282,0.0000278094,0.0001665953,0.00009034664,0.01342464],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6653277,"threshold_uncertainty_score":0.9995668,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1328243330302454,"score_gpt":0.3771460626394505,"score_spread":0.2443217296092051,"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."}}