{"id":"W3107198524","doi":"10.1109/cdc42340.2020.9303950","title":"Reinforcement Learning in Nonzero-sum Linear Quadratic Deep Structured Games: Global Convergence of Policy Optimization","year":2020,"lang":"en","type":"preprint","venue":"","topic":"Stochastic processes and financial applications","field":"Economics, Econometrics and Finance","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Reinforcement learning; Mathematical optimization; Mathematics; Convergence (economics); Nash equilibrium; Dimension (graph theory); Computer science; Mathematical economics; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":true,"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":[0.002609801,0.00139913,0.001555238,0.000557495,0.0004569199,0.001190314,0.001087921,0.001325147,0.002001985],"category_scores_gemma":[0.01073709,0.0005100009,0.0007444538,0.0004166164,0.002315962,0.001595065,0.002009768,0.001966712,0.0002420749],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001511665,"about_ca_system_score_gemma":0.001732081,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005978697,"about_ca_topic_score_gemma":0.003095686,"domain_scores_codex":[0.9990386,0.0004852221,0.00002921615,0.0001520191,0.0001511964,0.0001437666],"domain_scores_gemma":[0.9957383,0.0032579,0.0003504966,0.0001088529,0.0003176768,0.0002268252],"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":[0.00005152055,0.00004855951,0.0006106904,0.00007955419,0.00004129816,0.00007503162,0.00007882069,0.9236727,0.0007233865,0.06821179,0.0005748851,0.005831675],"study_design_scores_gemma":[0.000006452889,0.00001271742,0.00002868745,0.00000366664,0.000002752645,0.000004440241,0.000005950227,0.9885142,0.00007644343,0.01126035,0.00008190967,0.000002325447],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04440908,0.000312094,0.9498591,0.0006288614,0.00005541123,0.00005501615,0.00004087779,0.0001200423,0.00451947],"genre_scores_gemma":[0.9402611,0.0003302238,0.0544336,0.0002717582,0.00005494964,0.0001931468,0.0000761535,0.00008234676,0.004296818],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005978697,"threshold_uncertainty_score":0.01380211,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02577851295363042,"score_gpt":0.2580555533237466,"score_spread":0.2322770403701162,"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."}}