{"id":"W3038915804","doi":"10.48550/arxiv.2007.02151","title":"Variational Policy Gradient Method for Reinforcement Learning with General Utilities","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":37,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Reinforcement; Computer science; Gradient method; Mathematical optimization; Mathematical economics; Artificial intelligence; Mathematics; Engineering; Structural engineering","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002279163,0.001155578,0.0013633,0.0007138291,0.0004154415,0.001064149,0.001378943,0.001497113,0.004098907],"category_scores_gemma":[0.006133439,0.0006659971,0.0008306365,0.0005960364,0.001697663,0.00138121,0.001747697,0.002557037,0.0007279121],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002092689,"about_ca_system_score_gemma":0.002265319,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006985894,"about_ca_topic_score_gemma":0.004482653,"domain_scores_codex":[0.9993222,0.0003524457,0.00002332786,0.00009570161,0.0001489335,0.00005745983],"domain_scores_gemma":[0.9983898,0.001205623,0.0000732881,0.00007212545,0.0001891103,0.00007001888],"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":[0.00004344992,0.0000344514,0.0003307614,0.0001027243,0.0000420101,0.00006441939,0.00006727513,0.7629789,0.0008500533,0.2138607,0.001456431,0.02016882],"study_design_scores_gemma":[0.000005631646,0.000006160435,0.00001668912,0.000005391628,0.000002078345,0.000004037673,0.000002857176,0.976522,0.00009285843,0.02275494,0.0005847138,0.000002600829],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001884354,0.0002621269,0.9956354,0.000229417,0.00003898012,0.00003316391,0.00002191263,0.00009167508,0.001802964],"genre_scores_gemma":[0.4197453,0.001150483,0.5580292,0.0005078691,0.0001994105,0.0007445747,0.000293626,0.000535463,0.018794],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006985894,"threshold_uncertainty_score":0.01518357,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.084019057565281,"score_gpt":0.2294337399251543,"score_spread":0.1454146823598733,"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."}}