{"id":"W3032976898","doi":"10.48550/arxiv.2006.04551","title":"Cracking the Black Box: Distilling Deep Sports Analytics","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Sports Analytics and Performance","field":"Economics, Econometrics and Finance","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Computer science; Deep learning; Machine learning; Heuristics; Artificial intelligence; Analytics; Scalability; Artificial neural network; Black box; Deep neural networks; Tree (set theory); Big data; Data science; Transparency (behavior); Data mining; Database","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":[0.002481049,0.0009596696,0.0007804701,0.001019641,0.0004944536,0.002191626,0.001896308,0.001431022,0.002193416],"category_scores_gemma":[0.01619351,0.000635319,0.0004958518,0.001166465,0.001447479,0.007895566,0.004109875,0.003047444,0.0009074088],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001050485,"about_ca_system_score_gemma":0.001451808,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004670654,"about_ca_topic_score_gemma":0.00624795,"domain_scores_codex":[0.9990197,0.0003502335,0.00004477182,0.0002680313,0.000220539,0.00009660822],"domain_scores_gemma":[0.9959208,0.002509048,0.0003810544,0.0006912901,0.0002914521,0.000206273],"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":[0.0002944217,0.0002436124,0.009096486,0.0001761103,0.00008408362,0.000127688,0.0004872739,0.6844406,0.002619469,0.0545735,0.0062579,0.2415989],"study_design_scores_gemma":[0.000008178452,0.00002366531,0.0002285298,0.00001643075,0.000003945434,0.000006805374,0.00002954119,0.9367334,0.0006359598,0.06146283,0.0008435425,0.000007125023],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09151898,0.0007733949,0.899458,0.002357168,0.0000800553,0.00005702436,0.0005449089,0.001804474,0.003406017],"genre_scores_gemma":[0.8061687,0.0005324779,0.1892141,0.000650742,0.00009882877,0.0000939502,0.001077865,0.000252312,0.001911122],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004670654,"threshold_uncertainty_score":0.01312119,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1078925546982864,"score_gpt":0.1770439036347269,"score_spread":0.06915134893644041,"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."}}