{"id":"W4403925054","doi":"10.1214/24-aoas1933","title":"Learning risk preferences in Markov decision processes: An application to the fourth down decision in the national football league","year":2024,"lang":"en","type":"article","venue":"The Annals of Applied Statistics","topic":"Sports Analytics and Performance","field":"Economics, Econometrics and Finance","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; University of British Columbia","funders":"","keywords":"Football; League; Computer science; Artificial intelligence; Econometrics; Economics; Political science","routes":{"ca_aff":true,"ca_fund":false,"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":[0.01027225,0.001120844,0.002187325,0.00096261,0.0009180116,0.001976199,0.001532768,0.002570361,0.003912019],"category_scores_gemma":[0.03472149,0.0007664162,0.001765332,0.0008295485,0.002271711,0.001910952,0.002281977,0.004705676,0.0002276316],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002445932,"about_ca_system_score_gemma":0.002398427,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03097252,"about_ca_topic_score_gemma":0.01901261,"domain_scores_codex":[0.9969427,0.002029291,0.00008856249,0.0004530036,0.0001880073,0.0002984859],"domain_scores_gemma":[0.9594123,0.03660651,0.001947176,0.0005289057,0.0006788189,0.0008263168],"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":[0.000315545,0.0002475223,0.014007,0.00009444376,0.0001608841,0.0003551326,0.0005103002,0.8747351,0.0004289477,0.09502869,0.0008992883,0.01321713],"study_design_scores_gemma":[0.0000378088,0.00004292552,0.0009787686,0.00001428231,0.00002048433,0.00001822799,0.00007011694,0.9557018,0.00008111676,0.04275274,0.000259916,0.0000218287],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4280892,0.001540622,0.5606061,0.003919678,0.0001262909,0.0002076772,0.0003571471,0.0002524462,0.004900841],"genre_scores_gemma":[0.9606999,0.0007180047,0.03482672,0.0003458384,0.0001078659,0.0001512903,0.0002356043,0.00003772106,0.002876986],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03097252,"threshold_uncertainty_score":0.06158447,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0637202319053105,"score_gpt":0.3171072281836259,"score_spread":0.2533869962783154,"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."}}