{"id":"W3160261777","doi":"10.1016/j.automatica.2021.109693","title":"On the convergence of reinforcement learning with Monte Carlo Exploring Starts","year":2021,"lang":"en","type":"article","venue":"Automatica","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada; European Commission; Ontario Ministry of Research, Innovation and Science; Canada Research Chairs; Government of Canada; Pacific Institute for the Mathematical Sciences","keywords":"Reinforcement learning; Mathematical optimization; Convergence (economics); Bellman equation; Monte Carlo method; Computer science; Markov decision process; Complement (music); Function (biology); Mathematics; Markov process; Artificial intelligence; Economics","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.0002872181,0.0001316165,0.0001679348,0.00004505612,0.0001580727,0.00008870125,0.0005174458,0.00002311061,0.0001179024],"category_scores_gemma":[0.0002652293,0.0000888127,0.00004630129,0.0003554056,0.00005971401,0.0002811315,0.0002349566,0.0002052259,0.00006121305],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000380643,"about_ca_system_score_gemma":0.0001062546,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001299112,"about_ca_topic_score_gemma":0.000001471499,"domain_scores_codex":[0.9985366,0.000101677,0.000296349,0.0002230979,0.0005850117,0.0002572404],"domain_scores_gemma":[0.9984942,0.0004201957,0.0001807222,0.0007114796,0.00013923,0.00005414139],"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":[0.000003077171,0.000009692558,0.0003144058,0.00002930467,0.00003697941,0.00002097723,0.001975746,0.9243922,0.0001589668,0.0724264,0.0001467173,0.0004855625],"study_design_scores_gemma":[0.0001746368,0.000262665,0.0007835806,0.0001829261,0.00001040068,0.0000074264,0.0003528261,0.9884984,0.008727551,0.0001312011,0.0007356891,0.0001326947],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1626984,0.00002669777,0.8282521,0.001034141,0.0002158603,0.0001893332,1.268831e-7,0.0002061492,0.007377182],"genre_scores_gemma":[0.9889933,0.0000192896,0.009495622,0.0002028484,0.0000108704,0.0000293044,6.47799e-7,0.00001045028,0.001237707],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8262949,"threshold_uncertainty_score":0.3621677,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03576915530119511,"score_gpt":0.2220393188918922,"score_spread":0.1862701635906971,"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."}}