{"id":"W2787477569","doi":"10.1609/aaai.v32i1.11594","title":"PAC Reinforcement Learning With an Imperfect Model","year":2018,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"McGill University","keywords":"Reinforcement learning; Fidelity; Computer science; Reduction (mathematics); Action (physics); Simple (philosophy); Artificial intelligence; Sample (material); Transfer of learning; Imperfect; Sample complexity; State space; Polynomial; Machine learning; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002121028,0.001315705,0.002045392,0.0003725589,0.0005314507,0.001297574,0.002450725,0.001737912,0.002530509],"category_scores_gemma":[0.01105308,0.0007527206,0.000606444,0.0005183269,0.00222721,0.002906268,0.00189225,0.003791184,0.0004606357],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001656089,"about_ca_system_score_gemma":0.002317623,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00479539,"about_ca_topic_score_gemma":0.004883823,"domain_scores_codex":[0.9987025,0.0004808748,0.00005942439,0.0003247373,0.0002553478,0.0001770424],"domain_scores_gemma":[0.9915251,0.006141746,0.0005600767,0.001077771,0.0003700917,0.00032519],"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":[0.0001648391,0.00006582003,0.0004742221,0.00008660168,0.00003857628,0.0000984018,0.0000610417,0.9322212,0.0004106839,0.05089907,0.001233042,0.01424649],"study_design_scores_gemma":[0.00001978225,0.00002198937,0.00003480635,0.000003840673,0.000004663625,0.000009713055,0.000003656369,0.9791605,0.0001514287,0.02038874,0.0001970322,0.00000385422],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0291682,0.0001948484,0.9653115,0.0006133314,0.00005130838,0.00006492491,0.00009107451,0.0006298012,0.003875059],"genre_scores_gemma":[0.8438513,0.0001694716,0.149526,0.0003128973,0.00009374286,0.0002572969,0.0002209213,0.0001204531,0.005447965],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00479539,"threshold_uncertainty_score":0.01201582,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0664821800134268,"score_gpt":0.2958776474047536,"score_spread":0.2293954673913268,"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."}}