{"id":"W1787017840","doi":"10.1016/j.jmaa.2015.10.008","title":"Near optimality of quantized policies in stochastic control under weak continuity conditions","year":2015,"lang":"en","type":"preprint","venue":"Journal of Mathematical Analysis and Applications","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"Queen's University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Markov decision process; Discretization; Constructive; Markov process; Mathematical optimization; Optimal control; Stochastic control; Mathematics; Partially observable Markov decision process; Reduction (mathematics); Markov chain; Class (philosophy); Applied mathematics; Computer science; Process (computing); Mathematical analysis","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005665523,0.001375634,0.002855529,0.001528296,0.001050315,0.003457049,0.001978563,0.002815476,0.003938477],"category_scores_gemma":[0.02914317,0.001322605,0.001166811,0.001080215,0.005991375,0.005216358,0.004013832,0.004682667,0.000207853],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002904131,"about_ca_system_score_gemma":0.002902774,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003792975,"about_ca_topic_score_gemma":0.001871252,"domain_scores_codex":[0.9976292,0.001151245,0.0001366988,0.0003735824,0.0004421218,0.0002671342],"domain_scores_gemma":[0.9784765,0.01752629,0.001393933,0.0006109077,0.001125977,0.0008664344],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0002815122,0.000101643,0.0005812245,0.0002313317,0.00008092087,0.0001417248,0.0002394727,0.2932619,0.00161158,0.6961089,0.001017679,0.00634207],"study_design_scores_gemma":[0.00004773151,0.00007689364,0.0001815012,0.00003080699,0.00001293403,0.00002411519,0.00003931341,0.7325477,0.0003015815,0.2665015,0.0002177094,0.00001819047],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1558137,0.001112119,0.8259239,0.002803107,0.000223601,0.0001181203,0.0002296938,0.0002224614,0.01355338],"genre_scores_gemma":[0.963361,0.000685099,0.03054784,0.0003152641,0.0001486547,0.0001590317,0.0001345385,0.0001070502,0.004541661],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005665523,"threshold_uncertainty_score":0.02996254,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03001228517212436,"score_gpt":0.3246506552190392,"score_spread":0.2946383700469149,"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."}}