{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001298636,0.001008676,0.001043093,0.0003434738,0.0002443192,0.0007745373,0.0008510546,0.0009471307,0.001730642],"category_scores_gemma":[0.004253088,0.0004641925,0.0004570111,0.0003486314,0.0007701889,0.0008278429,0.001002353,0.001869935,0.0002431292],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001057687,"about_ca_system_score_gemma":0.001386138,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006258789,"about_ca_topic_score_gemma":0.005039108,"domain_scores_codex":[0.9996241,0.0001472192,0.00002020529,0.00007645905,0.00007250155,0.00005948881],"domain_scores_gemma":[0.9986022,0.0009768915,0.0001284781,0.00005978892,0.000161091,0.00007157311],"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":[0.00002747485,0.00002848031,0.0006196836,0.00002986852,0.00002617751,0.00002727885,0.0000150525,0.9812959,0.0003504809,0.004037513,0.0004542462,0.01308782],"study_design_scores_gemma":[0.000002749505,0.000006857169,0.00002887185,0.000002161836,0.000002116208,0.000002011284,9.859546e-7,0.9977928,0.00006187482,0.002028794,0.0000696258,0.000001151669],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05305006,0.001617459,0.9402828,0.0007993138,0.00009877382,0.00004016401,0.00009826,0.0006406636,0.003372387],"genre_scores_gemma":[0.952487,0.0004667458,0.04428132,0.0002132602,0.00005423588,0.00007999231,0.0001367159,0.00005324522,0.002227557],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006258789,"threshold_uncertainty_score":0.01244473,"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."}}