{"id":"W6929030644","doi":"10.48448/sa9g-t219","title":"Sample-Efficient Iterative Lower Bound Optimization of Deep Reactive Policies for Planning in Continuous MDPs","year":2022,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Upper and lower bounds; Gradient descent; Parametric statistics; Markov decision process; Artificial neural network; Optimization problem; Reinforcement learning; Range (aeronautics); Convergence (economics); Quality (philosophy)","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":[0.001215041,0.001298808,0.001047147,0.000415311,0.0003840859,0.0008808891,0.001132061,0.001115757,0.003416251],"category_scores_gemma":[0.0053832,0.0006701254,0.0005256928,0.0004045249,0.001168066,0.0009865966,0.001573448,0.0018842,0.000601684],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001312563,"about_ca_system_score_gemma":0.002551403,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01028283,"about_ca_topic_score_gemma":0.01125881,"domain_scores_codex":[0.9994855,0.000171029,0.00002348831,0.0001087176,0.0001318869,0.00007931645],"domain_scores_gemma":[0.9980861,0.00132611,0.000191139,0.0001343082,0.0001629651,0.00009939852],"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":[0.00003473833,0.00001954145,0.0001554873,0.00003540298,0.000009656573,0.00001728015,0.00001828976,0.9830638,0.0003269674,0.003671132,0.0005149583,0.01213278],"study_design_scores_gemma":[0.000004311702,0.000007546214,0.00001454399,0.000004056824,0.000001358219,0.000002469621,0.000002289244,0.9978619,0.0001617238,0.001781554,0.0001569986,0.0000011856],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01774141,0.0003089466,0.9766294,0.0002928682,0.00003342178,0.0000496522,0.00008417352,0.0008890025,0.003971057],"genre_scores_gemma":[0.7089465,0.0002866705,0.2854327,0.0002728072,0.00003867742,0.0002889585,0.0003513797,0.000416712,0.003965625],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01028283,"threshold_uncertainty_score":0.020446,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02553907224511928,"score_gpt":0.319526539367639,"score_spread":0.2939874671225198,"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."}}