{"id":"W1554366315","doi":"10.1023/a:1022145020786","title":"Approximate Gradient Methods in Policy-Space Optimization of Markov Reward Processes","year":2003,"lang":"en","type":"article","venue":"Discrete Event Dynamic Systems","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":52,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Convergence (economics); Variance (accounting); Markov chain; Computer science; Markov process; Gradient descent; Path (computing); Set (abstract data type); Mathematics; Process (computing); Mathematical optimization; Algorithm; Artificial intelligence; Statistics; Artificial neural network","routes":{"ca_aff":true,"ca_fund":false,"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":[0.001403634,0.0002268556,0.0003852737,0.0003893686,0.00006020597,0.0001068268,0.0006584824,0.00009310546,0.00000294395],"category_scores_gemma":[0.0006004188,0.0002076881,0.00007370704,0.001358536,0.0000494124,0.000349654,0.0001255042,0.0001367032,0.000003451212],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002625462,"about_ca_system_score_gemma":0.0002384951,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000156252,"about_ca_topic_score_gemma":0.000006859453,"domain_scores_codex":[0.9973853,0.0006198775,0.0007143887,0.0004290075,0.0004306875,0.0004207222],"domain_scores_gemma":[0.998437,0.0001660047,0.0004920283,0.0006750582,0.0001430186,0.0000868676],"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":[0.000003499345,0.00002301605,0.00101688,0.000437326,0.00002068611,0.000002879493,0.0008221722,0.9731426,0.00009012218,0.02424264,0.00001716355,0.0001809789],"study_design_scores_gemma":[0.000323111,0.00007683814,0.0001264845,0.0003126537,0.000007744342,0.00001841013,0.0002797845,0.9979719,0.0002009609,0.0001747211,0.0002872846,0.0002201256],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.000519559,0.0003928673,0.994657,0.000142821,0.0005365216,0.000613825,0.000002862724,0.00009454922,0.003039972],"genre_scores_gemma":[0.6605505,0.0001087105,0.3380757,0.00001619285,0.00001672406,0.00007923828,0.00001645473,0.00002808598,0.001108389],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.660031,"threshold_uncertainty_score":0.8469273,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01233715985053953,"score_gpt":0.3126734131687594,"score_spread":0.3003362533182199,"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."}}