{"id":"W2159272820","doi":"10.1613/jair.3175","title":"Non-Deterministic Policies in Markovian Decision Processes","year":2011,"lang":"en","type":"article","venue":"Journal of Artificial Intelligence Research","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"National Institutes of Health; Duke-NUS Medical School; Natural Sciences and Engineering Research Council of Canada; McGill University","keywords":"Reinforcement learning; Computer science; Flexibility (engineering); Markov decision process; Set (abstract data type); Task (project management); Action selection; Markov process; Construct (python library); Process (computing); Action (physics); Decision problem; Artificial intelligence; Selection (genetic algorithm); Decision support system; Machine learning; Operations research; Algorithm; Mathematics; Engineering","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":[0.003819138,0.0001272207,0.000249549,0.001219955,0.000153569,0.000291383,0.002078033,0.00008744935,0.00007762185],"category_scores_gemma":[0.002755599,0.0001050346,0.00007118322,0.001918238,0.0002285508,0.0007699018,0.0003740969,0.0007933617,0.0001521819],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001350784,"about_ca_system_score_gemma":0.0006723198,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009503392,"about_ca_topic_score_gemma":0.00006394683,"domain_scores_codex":[0.9967707,0.0001540909,0.0009914415,0.0002211468,0.001270898,0.0005917335],"domain_scores_gemma":[0.9971139,0.0008335555,0.0002763395,0.0003749777,0.001210637,0.0001905337],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008799643,0.00110402,0.01289345,0.0003415452,0.00006887745,0.001876284,0.04847098,0.08950875,0.005101085,0.05521543,0.0008626198,0.783677],"study_design_scores_gemma":[0.0002432036,0.00621681,0.01560253,0.00188499,0.00001588662,0.000489697,0.007659744,0.5799738,0.1576393,0.2281666,0.001230013,0.0008773938],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08328129,0.00007090092,0.9126673,0.0002223161,0.0003689837,0.0001577288,1.918376e-7,0.0000123674,0.003218895],"genre_scores_gemma":[0.9673474,0.0001780658,0.03220369,0.00002775236,0.0001334822,0.000003430591,8.985341e-8,0.00001073081,0.00009538501],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8840661,"threshold_uncertainty_score":0.4283186,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.238712003270168,"score_gpt":0.4267161564176883,"score_spread":0.1880041531475203,"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."}}