{"id":"W2753431380","doi":"","title":"Second-order Optimization 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":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Kronecker delta; Reinforcement learning; Computer science; Scalability; Curvature; Trust region; Mathematical optimization; Artificial intelligence; 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":[0.00152682,0.00119043,0.001200518,0.0004479228,0.0002770178,0.0009152578,0.001039582,0.001178769,0.002895904],"category_scores_gemma":[0.004582117,0.0005186946,0.0006644097,0.0003837667,0.001209941,0.001155321,0.001003516,0.001951764,0.0007713178],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001534535,"about_ca_system_score_gemma":0.001590851,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006630737,"about_ca_topic_score_gemma":0.005700902,"domain_scores_codex":[0.9995004,0.000170959,0.00002767039,0.00009431076,0.0001491568,0.00005753638],"domain_scores_gemma":[0.9987815,0.0006576595,0.0001025027,0.0001326043,0.0002481574,0.00007758602],"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":[0.00004530046,0.00002078862,0.0002869737,0.00003909355,0.00002487178,0.00003367279,0.00002918727,0.9666839,0.001184362,0.01230696,0.0007803082,0.0185647],"study_design_scores_gemma":[0.000002110792,0.000006874132,0.00001062696,0.000001809113,0.000001041556,0.000002898209,0.00000100604,0.9979782,0.0001423238,0.001726111,0.0001256921,0.000001403376],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004070576,0.0001350124,0.9944499,0.00007352127,0.00002556209,0.00001635136,0.00001675077,0.0003459661,0.0008664984],"genre_scores_gemma":[0.6943551,0.0003057997,0.2976253,0.0002009547,0.00005755537,0.0002084157,0.000180721,0.0004514969,0.006614653],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006630737,"threshold_uncertainty_score":0.01318431,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0371741992360738,"score_gpt":0.2844571715789954,"score_spread":0.2472829723429216,"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."}}