{"id":"W4386942299","doi":"10.48550/arxiv.2309.11404","title":"A representation-learning approach for insurance pricing with images","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Insurance and Financial Risk Management","field":"Economics, Econometrics and Finance","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Embedding; Computer science; Representation (politics); Portfolio; Unstructured data; Data science; Machine learning; Artificial intelligence; Data mining; Business; Big data; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001565959,0.0005562532,0.0008771293,0.001128673,0.0003482722,0.00172487,0.001991181,0.001694358,0.002239964],"category_scores_gemma":[0.005906646,0.0005405876,0.001116561,0.001343035,0.001071331,0.00271451,0.001419381,0.002695542,0.0005186858],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00107143,"about_ca_system_score_gemma":0.0006791445,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004544125,"about_ca_topic_score_gemma":0.003735525,"domain_scores_codex":[0.9993047,0.0002972438,0.00003115948,0.0001680094,0.0001244924,0.00007439158],"domain_scores_gemma":[0.9982692,0.00100946,0.0001951266,0.000257614,0.0001952394,0.00007329658],"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":[0.0001949749,0.0003295596,0.002676897,0.0001632275,0.0001404862,0.0001988476,0.0003347318,0.5959117,0.004064768,0.09872404,0.00584204,0.2914188],"study_design_scores_gemma":[0.000005811872,0.00001580578,0.0001560201,0.00000576423,0.000004613676,0.00001385547,0.00001422593,0.9750518,0.0002286241,0.02408622,0.0004114676,0.000005844431],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01673036,0.0003366953,0.9805362,0.0008674778,0.00004239079,0.00003768079,0.0001451227,0.000317264,0.0009868001],"genre_scores_gemma":[0.6219592,0.0006399284,0.3704928,0.0004415579,0.0003570582,0.0002000211,0.0008573541,0.0001405699,0.004911443],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004544125,"threshold_uncertainty_score":0.009035349,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1058802901486113,"score_gpt":0.1943896198261982,"score_spread":0.08850932967758687,"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."}}