{"id":"W2963313316","doi":"","title":"Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation","year":2017,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":92,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Kronecker delta; Reinforcement learning; Scalability; Computer science; Trust region; Curvature; Mathematical optimization; Artificial intelligence; Theoretical computer science; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0009493851,0.0002930807,0.0003414132,0.0002611706,0.002178416,0.004574608,0.001097786,0.000170013,0.000002295328],"category_scores_gemma":[0.0004749572,0.0002753902,0.0001023422,0.0001937076,0.00005320632,0.01248875,0.0002375589,0.0002700511,0.0000258228],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002536259,"about_ca_system_score_gemma":0.0001275106,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008821909,"about_ca_topic_score_gemma":5.932486e-7,"domain_scores_codex":[0.9974495,0.00009268441,0.000996241,0.0002987434,0.0006433509,0.0005195487],"domain_scores_gemma":[0.9963442,0.00007674344,0.00214865,0.0007243865,0.0005847944,0.0001212182],"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":[0.00001398555,0.000004738736,0.0003493462,0.0004647787,0.00001303959,4.8107e-7,0.001499256,0.9600219,0.0002023148,0.001997161,0.00006348728,0.03536959],"study_design_scores_gemma":[0.0007730199,0.000109847,0.0001218751,0.0001895573,0.00002014244,0.00004202838,0.0003055611,0.9927227,0.0008061146,0.00003936284,0.004550769,0.000318946],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001576075,0.00003949447,0.9940802,0.0001779397,0.0009905299,0.0009904318,3.772995e-7,0.0002776186,0.001867259],"genre_scores_gemma":[0.8919091,0.000003522369,0.1070055,0.0001001906,0.0001489677,0.00009959544,0.00004535049,0.00002080489,0.0006668938],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8903331,"threshold_uncertainty_score":0.9999698,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05248858218001681,"score_gpt":0.315826014867957,"score_spread":0.2633374326879402,"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."}}