{"id":"W2136723863","doi":"","title":"An Empirical Analysis of Off-policy Learning in Discrete MDPs","year":2012,"lang":"en","type":"article","venue":"European Workshop on Reinforcement Learning","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Reinforcement learning; Dynamic programming; Variance (accounting); Sampling (signal processing); Population; Mathematical optimization; Monte Carlo method; Policy analysis; Artificial intelligence; Algorithm; Statistics; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.009275687,0.0003096567,0.0006725016,0.00330378,0.0002980579,0.0002212232,0.001078291,0.00007730678,0.001321485],"category_scores_gemma":[0.008941299,0.0002473272,0.0003010324,0.007068608,0.0001790787,0.0008923978,0.0004516105,0.001308551,0.0006108911],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001921844,"about_ca_system_score_gemma":0.00006764719,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002156515,"about_ca_topic_score_gemma":0.000009795866,"domain_scores_codex":[0.9918799,0.002558089,0.001333447,0.0006662299,0.002532211,0.001030119],"domain_scores_gemma":[0.9960097,0.001864138,0.0005970396,0.0008874975,0.0002101729,0.0004314226],"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":[0.00009129043,0.00005564754,0.09694895,0.000002139311,0.00008571392,0.00001279352,0.002584395,0.7895367,0.00008266271,0.0001340315,0.00008505394,0.1103806],"study_design_scores_gemma":[0.0008099185,0.0004431375,0.2598178,0.00007190767,0.00007644366,0.000001750352,0.005164592,0.6482854,0.0001233031,0.00002449435,0.08470932,0.0004718919],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6798519,0.0002854003,0.2291816,0.0006455171,0.0002241845,0.0004874186,0.00000298883,0.0002283111,0.08909265],"genre_scores_gemma":[0.9837555,0.00009230338,0.0004232932,0.0001766651,0.0002768363,0.000006487121,0.00004720776,0.00005123543,0.0151705],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3039035,"threshold_uncertainty_score":0.9999979,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1071006399021413,"score_gpt":0.4590256981712578,"score_spread":0.3519250582691165,"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."}}