{"id":"W2941011530","doi":"","title":"Reward Estimation for Variance Reduction in Deep Reinforcement Learning","year":2018,"lang":"en","type":"article","venue":"International Conference on Learning Representations","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"","keywords":"Reinforcement learning; Variance reduction; Reduction (mathematics); Computer science; Variance (accounting); Estimation; Artificial intelligence; Reinforcement; Machine learning; Statistics; Econometrics; Psychology; Mathematics; Engineering; Economics; Social psychology","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":[0.0005266555,0.0001809399,0.0001572973,0.0004426208,0.0003647114,0.000399239,0.0007263854,0.00008665099,0.0001946417],"category_scores_gemma":[0.001338798,0.0002043368,0.00007214658,0.0004178937,0.00009674174,0.0008841581,0.0001484517,0.0004345335,0.0002071008],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000233784,"about_ca_system_score_gemma":0.0001139015,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006449973,"about_ca_topic_score_gemma":0.000007224717,"domain_scores_codex":[0.9979142,0.0001416655,0.0005085344,0.0005634328,0.0005555133,0.0003166423],"domain_scores_gemma":[0.9983661,0.0002352078,0.0003558306,0.0003513254,0.0006206834,0.00007085899],"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":[0.00003592214,0.00002251442,0.0009021045,0.000006721565,0.00002240631,0.000001680698,0.002173597,0.8247232,0.0006026715,0.1642685,0.0001000622,0.007140503],"study_design_scores_gemma":[0.0006206697,0.0003707816,0.001445003,0.00009933899,0.000005736475,0.00001010188,0.0003509843,0.9904602,0.0007609829,0.003289272,0.002384313,0.000202627],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002523527,0.000003831643,0.9629498,0.00233649,0.00103834,0.000390612,3.513616e-7,0.0002206092,0.03053643],"genre_scores_gemma":[0.9528331,0.00002912466,0.03864229,0.00008568347,0.0002273258,0.0001618394,0.00008436644,0.00001744572,0.007918783],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9503096,"threshold_uncertainty_score":0.8332614,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.050061910459264,"score_gpt":0.3514664291660533,"score_spread":0.3014045187067893,"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."}}