{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01100739,0.002098893,0.003125587,0.00132453,0.001175502,0.003009299,0.003633662,0.003295624,0.00596545],"category_scores_gemma":[0.07236342,0.001287959,0.001991492,0.001179243,0.004195905,0.008292516,0.004028942,0.007149305,0.0005505772],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005132472,"about_ca_system_score_gemma":0.004838794,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005583488,"about_ca_topic_score_gemma":0.005626652,"domain_scores_codex":[0.9943037,0.002547208,0.0003159191,0.001108223,0.001081,0.0006439975],"domain_scores_gemma":[0.8550571,0.1332763,0.003566022,0.0044954,0.00200553,0.001599735],"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":[0.000932378,0.0003693839,0.003459416,0.0004452734,0.0002432898,0.0001319127,0.0001508075,0.8774789,0.001324134,0.09177452,0.002400396,0.02128958],"study_design_scores_gemma":[0.00004093758,0.00007093714,0.0002376116,0.00002589973,0.00001747118,0.00001900961,0.00001383143,0.944951,0.000358109,0.0540437,0.0002088397,0.0000126506],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09166767,0.002896925,0.891506,0.004561751,0.0002068173,0.0003863686,0.0008662193,0.0009205999,0.006987596],"genre_scores_gemma":[0.8127196,0.001784523,0.1758021,0.001366898,0.0005157401,0.001181546,0.001727254,0.0005580306,0.004344291],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01100739,"threshold_uncertainty_score":0.05821335,"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."}}