{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003990834,0.0003453649,0.0005967357,0.0002050284,0.0002327227,0.0001720208,0.0008637212,0.0002528623,0.0005747197],"category_scores_gemma":[0.00005067973,0.0003647443,0.0003937018,0.0005190258,0.0001828916,0.0001519498,0.0006383533,0.0007880311,0.000329017],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001763868,"about_ca_system_score_gemma":0.00005466793,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002478121,"about_ca_topic_score_gemma":0.00007034807,"domain_scores_codex":[0.9980288,0.00001118597,0.0005394064,0.0009855424,0.00004360748,0.0003915276],"domain_scores_gemma":[0.9979391,0.00004878922,0.0008204915,0.0009519832,0.00007078316,0.0001688452],"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":[0.00001853108,0.00003650043,0.1321987,0.0000844778,0.000160325,0.0001861328,0.0003165894,0.602937,2.825086e-7,0.2636057,0.0003361438,0.0001195426],"study_design_scores_gemma":[0.0001839652,0.00001546638,0.0165332,0.00004925736,0.00009504005,0.000001395349,0.0001616755,0.8837438,0.000003529372,0.07811157,0.020611,0.0004901648],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7077553,0.0008867081,0.2471819,0.0009732847,0.001355575,0.0004518412,0.0002228667,0.0001467075,0.0410258],"genre_scores_gemma":[0.9944127,0.002034377,0.00008049748,0.0003449598,0.0003160916,5.154185e-7,0.00006088844,0.00004453811,0.002705453],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2866574,"threshold_uncertainty_score":0.9998804,"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."}}