{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006178284,0.0002858968,0.0002542661,0.0001423376,0.0004342513,0.0004012135,0.002322532,0.00008875108,0.00006357094],"category_scores_gemma":[0.0002529796,0.000195893,0.00007799167,0.0006369992,0.0004689776,0.0008655668,0.0004624169,0.0004461812,0.0001135665],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006588811,"about_ca_system_score_gemma":0.0001610318,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002647506,"about_ca_topic_score_gemma":0.000006147488,"domain_scores_codex":[0.9976273,0.00002234925,0.0005101647,0.0005579037,0.0007863495,0.0004959923],"domain_scores_gemma":[0.997844,0.00005032175,0.0004847532,0.0004785731,0.001005522,0.0001367615],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001091581,0.00004662146,0.0002238433,0.00002149492,0.00001920396,3.255154e-7,0.002582887,0.3301497,0.01324819,0.6426756,0.00004055259,0.01088245],"study_design_scores_gemma":[0.00003037527,0.001582703,0.00002609066,0.0001256846,0.00001027001,0.000003739119,0.000281171,0.7899683,0.1971015,0.01061051,0.0000431454,0.0002165],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07177303,0.000002733367,0.8979242,0.0007518049,0.0001885369,0.0003874348,3.12879e-7,0.0001916999,0.02878026],"genre_scores_gemma":[0.9857633,0.00001183083,0.01281446,0.0002187983,0.00008696041,0.00001911781,5.425423e-7,0.00002012807,0.001064847],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9139903,"threshold_uncertainty_score":0.7988285,"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."}}