{"id":"W4411535800","doi":"10.3390/jrfm18070347","title":"Risk-Sensitive Deep Reinforcement Learning for Portfolio Optimization","year":2025,"lang":"en","type":"article","venue":"Journal of risk and financial management","topic":"Market Dynamics and Volatility","field":"Economics, Econometrics and Finance","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National University of Singapore","keywords":"Reinforcement learning; Sharpe ratio; Futures contract; Volatility (finance); Portfolio optimization; Portfolio; Asset allocation; Computer science; Artificial intelligence; Economics; Financial economics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001235831,0.0001039721,0.0003035792,0.0003281712,0.0002040687,0.00005381565,0.00008199741,0.00005825803,0.0000269387],"category_scores_gemma":[0.0003399829,0.000109821,0.0001306456,0.0001780561,0.0000230103,0.0001193682,0.00005940449,0.0001774329,0.000001009294],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007135195,"about_ca_system_score_gemma":0.00001181693,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004234602,"about_ca_topic_score_gemma":0.0000108316,"domain_scores_codex":[0.998957,0.00001996928,0.0006647702,0.0001663246,0.00003746954,0.0001544015],"domain_scores_gemma":[0.9988527,0.00006893676,0.0008503055,0.00009710268,0.00008908501,0.00004187837],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006474273,0.0001313206,0.2994624,0.000178411,0.0002206567,0.00001387154,0.0005034719,0.2731728,2.725847e-7,0.2440474,0.001019755,0.1806023],"study_design_scores_gemma":[0.00166409,0.000204689,0.1472006,0.00005280682,0.0001055688,0.000001617929,0.0001734739,0.7199785,0.000002195214,0.04632358,0.0841113,0.0001815354],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0249499,0.0006627013,0.9631298,0.00005573426,0.0004254336,0.0002367259,0.00001030934,0.000005521637,0.01052387],"genre_scores_gemma":[0.9816245,0.00779654,0.009523742,0.00008365418,0.00008411156,0.000007562205,0.000006160471,0.000007523749,0.0008662312],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9566746,"threshold_uncertainty_score":0.4478372,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006643500312059729,"score_gpt":0.2044658305040042,"score_spread":0.1978223301919444,"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."}}