{"id":"W191658262","doi":"","title":"{Toward Minimax Off-policy Value Estimation}","year":2015,"lang":"en","type":"article","venue":"","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":62,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Estimator; Minimax; Markov decision process; Multiplicative function; Mathematical optimization; Oracle; Computer science; Time horizon; Limit (mathematics); Sample size determination; Upper and lower bounds; Mathematics; Sample (material); Value (mathematics); Markov process; 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":["metaresearch","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00266786,0.0001307192,0.0002328663,0.0005472691,0.00009107888,0.0002943794,0.001006016,0.00007045666,0.0007330045],"category_scores_gemma":[0.0204206,0.00008762778,0.00007521295,0.001659563,0.0001799056,0.0007037286,0.0003180579,0.0001495448,0.008909793],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001795816,"about_ca_system_score_gemma":0.0006437635,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001221741,"about_ca_topic_score_gemma":0.00001111738,"domain_scores_codex":[0.9955585,0.0001737774,0.0004866736,0.0004562957,0.002918824,0.0004059198],"domain_scores_gemma":[0.9969017,0.001049589,0.00009990436,0.0007090211,0.0007922433,0.0004475654],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00005704639,0.0001085507,0.0005912926,0.000003161584,0.00001558,0.0000441661,0.001443675,0.03552714,0.00004900308,0.02406199,0.1869234,0.751175],"study_design_scores_gemma":[0.0008094794,0.0001387395,0.001379533,0.000004355366,0.000002600212,0.00003732887,0.00136456,0.3852173,0.001313905,0.4326744,0.1768337,0.0002240616],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006838641,0.0001835381,0.868857,0.0109324,0.0004132479,0.0003005927,0.0000149082,0.0001698037,0.1122899],"genre_scores_gemma":[0.7132871,0.0000169153,0.1951555,0.0009562553,0.0006839443,0.00003925688,0.000008522326,0.00003462254,0.0898179],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.750951,"threshold_uncertainty_score":0.9918619,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3651976247050544,"score_gpt":0.5253249721812989,"score_spread":0.1601273474762445,"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."}}