{"id":"W2996026485","doi":"10.48550/arxiv.1912.06875","title":"Natural Actor-Critic Converges Globally for Hierarchical Linear Quadratic Regulator","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Booth University College","funders":"","keywords":"Curse of dimensionality; Reinforcement learning; Linear-quadratic regulator; Convergence (economics); Mathematical optimization; State space; Quadratic equation; Computer science; Action (physics); Mathematics; Optimal control; Artificial intelligence","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":[0.001907858,0.0009470129,0.001151605,0.0004201526,0.0004275994,0.0007696291,0.0008971244,0.00119412,0.002653335],"category_scores_gemma":[0.006394231,0.0003830164,0.0005278443,0.0003318062,0.001351152,0.0008551401,0.001576474,0.001587511,0.0004675118],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001241252,"about_ca_system_score_gemma":0.001397495,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004227535,"about_ca_topic_score_gemma":0.003607008,"domain_scores_codex":[0.9993522,0.0002280682,0.00002843824,0.0001672956,0.0001474029,0.00007668993],"domain_scores_gemma":[0.9974006,0.001818288,0.0002361141,0.000130061,0.0003006204,0.000114298],"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":[0.00003646217,0.00003146361,0.0003999546,0.00009631731,0.00002787051,0.00006248776,0.00008581888,0.9355814,0.001288185,0.0498046,0.001221263,0.01136421],"study_design_scores_gemma":[0.000005029638,0.00001133374,0.00003117345,0.000003509479,0.00000171133,0.000004676657,0.000003924987,0.9903673,0.0000839285,0.009320157,0.0001652742,0.000001984108],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01609633,0.0002214532,0.9768287,0.0004032343,0.00003632673,0.00004363526,0.00003749789,0.0003254034,0.006007458],"genre_scores_gemma":[0.8820672,0.0002419443,0.1104679,0.0002754293,0.00005050698,0.0002730876,0.0001325824,0.0001465824,0.006344798],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004227535,"threshold_uncertainty_score":0.01008981,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05382236452644264,"score_gpt":0.2089326983325367,"score_spread":0.1551103338060941,"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."}}