{"id":"W2945374007","doi":"10.1007/978-3-030-18305-9_68","title":"Safe Policy Learning with Constrained Return Variance","year":2019,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Computer science; Variance (accounting); Artificial intelligence; Machine learning; Accounting","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":[0.001914524,0.001151084,0.001513337,0.0006215898,0.0004531116,0.001530388,0.001622264,0.001840862,0.004959082],"category_scores_gemma":[0.007386802,0.0008825029,0.0007379969,0.0007610809,0.001824896,0.002052321,0.002929731,0.003278645,0.001309955],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001105272,"about_ca_system_score_gemma":0.001734065,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001922567,"about_ca_topic_score_gemma":0.001396904,"domain_scores_codex":[0.9989586,0.0003079871,0.00004837355,0.0002016382,0.0003438991,0.0001394726],"domain_scores_gemma":[0.9968904,0.002287359,0.0001549117,0.0003199109,0.0002222055,0.0001252028],"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":[0.0001560853,0.00005924798,0.000247868,0.0001025025,0.00006029411,0.00008898433,0.00007258308,0.6955435,0.001814752,0.1763655,0.002940051,0.1225487],"study_design_scores_gemma":[0.00001555234,0.00003000921,0.00003869163,0.00001329855,0.000007067469,0.00001986388,0.000003879514,0.8799238,0.0007114387,0.1183911,0.000837532,0.000007830498],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003745821,0.0002390463,0.9924932,0.0001619835,0.00005288815,0.00001466454,0.00002767944,0.0003747368,0.002889927],"genre_scores_gemma":[0.6025313,0.001016682,0.3637662,0.0004492188,0.0002686558,0.0002706297,0.0003370082,0.0007347132,0.03062575],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004959082,"threshold_uncertainty_score":0.01658976,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01184008478179542,"score_gpt":0.2357360834831554,"score_spread":0.2238959987013599,"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."}}