{"id":"W3212744543","doi":"10.48550/arxiv.2111.08172","title":"Off-Policy Actor-Critic with Emphatic Weightings","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Counterexample; Variety (cybernetics); Computer science; Variance (accounting); Work (physics); Gradient method; Mathematical optimization; Mathematics; Mathematical economics; Applied mathematics; Economics; Artificial intelligence; Discrete mathematics","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":[0.002429823,0.001178867,0.000979299,0.0003685213,0.0003257018,0.001157199,0.00125168,0.001107615,0.002581285],"category_scores_gemma":[0.009372066,0.0005675409,0.0003901567,0.0003459631,0.001309658,0.00144607,0.001355968,0.002023945,0.0006504048],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009329986,"about_ca_system_score_gemma":0.001487124,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002352051,"about_ca_topic_score_gemma":0.002239071,"domain_scores_codex":[0.9990022,0.0004160617,0.00005203138,0.0001833266,0.0002414116,0.0001050714],"domain_scores_gemma":[0.9972833,0.001696437,0.0002498357,0.000286552,0.0003748748,0.0001090083],"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":[0.0001506657,0.00007241113,0.00076761,0.0001076418,0.00004008124,0.00008127973,0.0001000348,0.887822,0.002545745,0.04420405,0.001467531,0.06264085],"study_design_scores_gemma":[0.00001138948,0.00001946194,0.00005113241,0.000007882728,0.000004285684,0.00001244106,0.000006096624,0.9917128,0.000628742,0.007111598,0.000430318,0.000003790469],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01393412,0.0001239407,0.9811383,0.000216354,0.00003981972,0.00005326349,0.00001961786,0.0004065192,0.004067998],"genre_scores_gemma":[0.7898219,0.0001785644,0.2031463,0.0003263623,0.00004217459,0.0001585443,0.00007614978,0.0001712197,0.006078681],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002581285,"threshold_uncertainty_score":0.01285028,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04648557504175391,"score_gpt":0.186150771689604,"score_spread":0.1396651966478501,"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."}}