{"id":"W4417094986","doi":"10.48550/arxiv.2502.16772","title":"Model-Based Exploration in Monitored Markov Decision Processes","year":2025,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Alberta Innovates; University of Alberta; Alberta Machine Intelligence Institute; Alliance de recherche numérique du Canada; Mitacs; Canadian Institute for Advanced Research","keywords":"Markov decision process; Leverage (statistics); Exploit; Partially observable Markov decision process; Convergence (economics); Reinforcement learning; Process (computing); Markov process","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":[0.002472923,0.0008534961,0.001253164,0.0005243384,0.0004066745,0.001132745,0.00123706,0.001047163,0.001691563],"category_scores_gemma":[0.01112184,0.0007178119,0.0006662536,0.0005403913,0.001517975,0.001729397,0.001458864,0.001938074,0.0001909484],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001541924,"about_ca_system_score_gemma":0.001621196,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004830432,"about_ca_topic_score_gemma":0.003809011,"domain_scores_codex":[0.9989002,0.0005447844,0.00004474545,0.0002125995,0.0001660635,0.0001315892],"domain_scores_gemma":[0.9923654,0.006260115,0.0005991358,0.0003063532,0.0002536966,0.0002151156],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00005352304,0.00002093236,0.0005290501,0.00004188598,0.00001885019,0.00003324682,0.00004544465,0.9680734,0.0001974574,0.02378758,0.0002099513,0.006988602],"study_design_scores_gemma":[0.000008917101,0.0000127941,0.00004131023,0.00000544895,0.000002662834,0.000005735195,0.000003607796,0.9817035,0.00009652969,0.01800137,0.0001155716,0.000002599996],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03805291,0.0003823687,0.9587674,0.0003586711,0.00002217745,0.00004140452,0.00007482548,0.0003590664,0.001941235],"genre_scores_gemma":[0.9139949,0.0002917465,0.08374949,0.0001082493,0.00002500553,0.0001711555,0.0001154153,0.00005771869,0.001486243],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004830432,"threshold_uncertainty_score":0.01307821,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09140309050084275,"score_gpt":0.2201448655673335,"score_spread":0.1287417750664907,"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."}}