{"id":"W4394658390","doi":"10.66573/001c.144283","title":"Synthesizing Property & Casualty Ratemaking Datasets using Generative Adversarial Networks","year":2025,"lang":"en","type":"preprint","venue":"Variance","topic":"Autopsy Techniques and Outcomes","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"University of Waterloo; Alliance de recherche numérique du Canada; Brigham Young University","keywords":"Categorical variable; Computer science; Generative grammar; Confidentiality; Data mining; Differential privacy; Generative adversarial network; Property (philosophy); Adversarial system; Transparency (behavior); Generative model; Machine learning; Artificial intelligence; Deep learning; Computer security","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005659961,0.0004711424,0.000932659,0.0001348522,0.0002266524,0.00009717535,0.000324184,0.0006646639,0.000118102],"category_scores_gemma":[0.0004035293,0.0003385404,0.0002086412,0.0002132004,0.00007679355,0.00009931095,0.0009592483,0.001261947,0.000007833175],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004049107,"about_ca_system_score_gemma":0.0009585597,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009767658,"about_ca_topic_score_gemma":0.00009257457,"domain_scores_codex":[0.9975916,0.00018881,0.0005743471,0.0008724768,0.000316971,0.0004558355],"domain_scores_gemma":[0.9980376,0.0001479561,0.0003864569,0.001163805,0.0001475721,0.000116655],"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":[0.007045246,0.002780649,0.03761174,0.02914035,0.0114462,0.006886652,0.008909084,0.4301414,0.02219462,0.03196466,0.2766415,0.1352378],"study_design_scores_gemma":[0.002838061,0.0001461527,0.002438649,0.02328577,0.002819885,0.0003538312,0.0001034468,0.6455035,0.004684949,0.001516837,0.31359,0.002718908],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002315528,0.001685062,0.9843515,0.0009818996,0.002941319,0.002297672,0.0007350649,0.0005382781,0.004153725],"genre_scores_gemma":[0.235858,0.001268127,0.7325476,0.005884381,0.005782478,0.000346294,0.003405683,0.000207446,0.01469991],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2518038,"threshold_uncertainty_score":0.9999067,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08169857589428087,"score_gpt":0.3641907366714952,"score_spread":0.2824921607772143,"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."}}