{"id":"W3177464508","doi":"","title":"On the Sample Complexity of Batch Reinforcement Learning with Policy-Induced Data.","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Omega; Sample complexity; Horizon; Markov decision process; Upper and lower bounds; Time horizon; Function (biology); Exponential function; Combinatorics; Mathematics; Sample (material); Finite set; Binary logarithm; Markov process; Mathematical optimization; Physics; Computer science; Statistics; Artificial intelligence; Mathematical analysis; Quantum mechanics","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"],"consensus_categories":[],"category_scores_codex":[0.0005658471,0.0003741085,0.0004371284,0.0002395377,0.0003120287,0.0002185644,0.00462987,0.0001775959,0.00008359509],"category_scores_gemma":[0.0004410854,0.0003157268,0.00012623,0.0009466784,0.0002760098,0.0003671225,0.006742943,0.001211907,0.0000215411],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000256478,"about_ca_system_score_gemma":0.0007131326,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001783656,"about_ca_topic_score_gemma":0.00007829229,"domain_scores_codex":[0.9974472,0.0003377158,0.0003228466,0.001106741,0.0003451041,0.0004404293],"domain_scores_gemma":[0.9943863,0.000682585,0.0006573736,0.00386925,0.0002778558,0.000126696],"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":[0.00002151375,0.00002466563,0.0005463802,0.00004608541,0.0001245031,0.00003242587,0.0003060162,0.7372183,0.00001880531,0.2615461,0.00005761764,0.00005752236],"study_design_scores_gemma":[0.000367147,0.0002358721,0.000560672,0.0002253786,0.00005310428,0.000002594716,0.0002906253,0.9944819,0.0002846579,0.002911526,0.0002246671,0.0003618659],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.087876,0.000004784752,0.9072679,0.0004172509,0.0001693128,0.0003502288,0.000006438012,0.0001299689,0.003778165],"genre_scores_gemma":[0.9947361,0.00004469818,0.004012981,0.0001940611,0.00004962997,8.897345e-7,0.0001492198,0.0000224507,0.0007899732],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9068601,"threshold_uncertainty_score":0.9999295,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2023558664238282,"score_gpt":0.2316189306737228,"score_spread":0.02926306424989455,"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."}}