{"id":"W2963582321","doi":"","title":"Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning","year":2017,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":103,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Regret; Reinforcement learning; Computer science; Bridge (graph theory); State (computer science); Mathematical optimization; Algorithm; Artificial intelligence; Machine learning; Mathematics","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":["sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.003263447,0.000222998,0.0003641992,0.0004068513,0.003243982,0.006460275,0.0007948293,0.0001296414,0.00001407121],"category_scores_gemma":[0.004696167,0.0001655907,0.00007012072,0.0002666835,0.0002491075,0.009944907,0.0002445624,0.0003219703,0.00005600717],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000128363,"about_ca_system_score_gemma":0.000144295,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006792674,"about_ca_topic_score_gemma":0.000006990971,"domain_scores_codex":[0.9964268,0.00007328281,0.001137829,0.000298318,0.001547526,0.0005162232],"domain_scores_gemma":[0.9961886,0.0003914144,0.001526036,0.0005954858,0.001113852,0.0001846034],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000178985,0.00001238203,0.004662934,0.0008027201,0.00002379052,0.000004207585,0.005301728,0.1096131,0.0001122893,0.001692256,0.00181715,0.8757784],"study_design_scores_gemma":[0.0008668543,0.0001652092,0.001114932,0.0001703778,0.0000067305,0.00004062248,0.003804845,0.9007249,0.0001508037,0.0009675877,0.09174553,0.0002415661],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06905583,0.0008329116,0.9094979,0.002391006,0.001970556,0.002618249,0.00003095617,0.0003608538,0.0132417],"genre_scores_gemma":[0.9941895,0.00002170264,0.0005779187,0.00007812714,0.0001754356,0.0001094627,0.00002942583,0.00001574375,0.004802662],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9251337,"threshold_uncertainty_score":0.9980537,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1348757675199226,"score_gpt":0.4272213443219531,"score_spread":0.2923455768020305,"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."}}