{"id":"W3105702366","doi":"","title":"Variational Policy Gradient Method for Reinforcement Learning with General Utilities","year":2020,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Mathematical optimization; Reinforcement learning; Convexity; Markov decision process; Q-learning; Mathematics; Gradient descent; Gradient method; Computer science; Applied mathematics; Convergence (economics); Markov process; Artificial neural network; Artificial intelligence; Finance","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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003908071,0.0002146866,0.0002306061,0.0001860984,0.0004415797,0.001134628,0.0004852545,0.00006590584,0.000002857777],"category_scores_gemma":[0.000204691,0.0001779362,0.00005660672,0.0005418209,0.0000274789,0.003710137,0.0001059795,0.0002017507,0.0000191331],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000128507,"about_ca_system_score_gemma":0.0002931137,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005538425,"about_ca_topic_score_gemma":2.155502e-7,"domain_scores_codex":[0.9980273,0.00007504048,0.0006796217,0.0002159874,0.0006268821,0.0003751874],"domain_scores_gemma":[0.9985284,0.00008740531,0.0006096754,0.0001774994,0.0004458071,0.0001511753],"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":[0.00002077496,0.000002385714,0.00007636803,0.00035227,0.00001574516,3.097567e-7,0.005143465,0.9296376,0.00002774658,0.05874804,0.0003705171,0.005604844],"study_design_scores_gemma":[0.0005867178,0.0003543159,0.00005795512,0.00004915728,0.000008537547,0.00002212519,0.0003968434,0.9685346,0.0001146451,0.0000329829,0.02963061,0.0002114814],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0001841787,0.00002758901,0.9940392,0.002340293,0.0002150182,0.0005975977,0.000001785294,0.0003887546,0.00220558],"genre_scores_gemma":[0.8497978,0.000003015467,0.1456425,0.002743348,0.0004729659,0.0002205666,0.0001162006,0.00001974823,0.0009838685],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8496136,"threshold_uncertainty_score":0.9999023,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03067472875020712,"score_gpt":0.2824091564659542,"score_spread":0.2517344277157471,"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."}}