{"id":"W4416052765","doi":"10.48550/arxiv.2504.11412","title":"Measures of Variability for Risk-averse Policy Gradient","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Risk and Portfolio Optimization","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alliance de recherche numérique du Canada; Government of Canada; Canadian Institute for Advanced Research","keywords":"CVAR; Standard deviation; Randomness; Gini coefficient; Quantile; Work (physics); Expected shortfall; Empirical research","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.007407903,0.0002836026,0.0006916772,0.0005955279,0.0001649779,0.00007670966,0.001138772,0.0003738634,0.00009141268],"category_scores_gemma":[0.02226025,0.0002247829,0.0005199037,0.0007934921,0.0001580127,0.0001122132,0.0006399526,0.000358828,0.00004256619],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001155173,"about_ca_system_score_gemma":0.0008099096,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001512936,"about_ca_topic_score_gemma":0.0001239428,"domain_scores_codex":[0.9959754,0.0006479073,0.001230745,0.0009394459,0.0009026057,0.0003038523],"domain_scores_gemma":[0.993799,0.002027496,0.001154321,0.001780092,0.001109319,0.0001297879],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00009793615,0.0001396142,0.9322637,0.0000426888,0.00008644963,6.546971e-7,0.0007693375,0.02797803,0.00001705941,0.00277587,0.006364957,0.02946369],"study_design_scores_gemma":[0.0008114838,0.0001030179,0.6536523,0.0001054375,0.0002611684,0.000001038545,0.0002369408,0.01403312,0.001569758,0.2201557,0.1085237,0.0005463273],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7722212,0.0002073268,0.2086019,0.0009007091,0.002573765,0.00150907,0.001163901,0.00009120575,0.01273089],"genre_scores_gemma":[0.9903157,0.001100137,0.0047202,0.0001152746,0.0003071169,0.00009449575,0.00005342099,0.00001432293,0.003279276],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2786115,"threshold_uncertainty_score":0.9859757,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1963626164584548,"score_gpt":0.4111475165118763,"score_spread":0.2147849000534215,"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."}}