{"id":"W4225482307","doi":"10.1111/mafi.12388","title":"Reinforcement learning with dynamic convex risk measures","year":2023,"lang":"en","type":"article","venue":"Mathematical Finance","topic":"Risk and Portfolio Optimization","field":"Decision Sciences","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Reinforcement learning; Computer science; Mathematical optimization; Flexibility (engineering); Q-learning; Dynamic programming; Artificial neural network; Obstacle; Dynamic risk measure; Convex optimization; Artificial intelligence; Value at risk; Regular polygon; Risk management; Mathematics; Economics; Finance","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":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001853607,0.0001227829,0.0002594003,0.0001396785,0.0002129481,0.000129998,0.0003062102,0.00005362554,0.0001754606],"category_scores_gemma":[0.002097243,0.00007627893,0.00005762152,0.001080856,0.00009231528,0.0001619306,0.00006540502,0.000168291,0.003759647],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001845383,"about_ca_system_score_gemma":0.00003517628,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004540199,"about_ca_topic_score_gemma":0.000003591401,"domain_scores_codex":[0.9977933,0.0000931364,0.000454539,0.0003028981,0.0010911,0.0002650543],"domain_scores_gemma":[0.9985083,0.0006494469,0.0002338262,0.0003971859,0.0001557971,0.00005545689],"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":[0.00007017187,0.00004595091,0.009729676,0.00001149157,0.00002482911,0.0000397492,0.001870477,0.7935531,0.00003830041,0.04246737,0.00517802,0.1469709],"study_design_scores_gemma":[0.0003087725,0.0001325671,0.008221028,0.0000454899,0.00001482867,0.000008856328,0.0002829673,0.8622267,0.0001070847,0.08223691,0.04621929,0.0001954957],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2550234,0.00005792941,0.7224036,0.0005033303,0.00009631117,0.0002714565,0.000002894688,0.0002203029,0.0214208],"genre_scores_gemma":[0.9729105,0.0008718638,0.004456731,0.00003442944,0.0000163054,0.00002794331,0.000005147719,0.00001450418,0.02166261],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7179468,"threshold_uncertainty_score":0.997016,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04896239240589748,"score_gpt":0.3373794269646531,"score_spread":0.2884170345587557,"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."}}