{"id":"W3209416948","doi":"10.48550/arxiv.2111.00062","title":"Improving Generalization Bounds for VC Classes Using the Hypergeometric Tail Inversion","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa; Université Laval","funders":"","keywords":"Hypergeometric distribution; Upper and lower bounds; Inversion (geology); Generalization; Mathematics; Margin (machine learning); Hypergeometric function; Combinatorics; Set (abstract data type); Algorithm; Discrete mathematics; Computer science; Statistics; Pure mathematics; Mathematical analysis; Machine learning","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":[0.01548253,0.002781556,0.00279378,0.004287518,0.001560964,0.005292586,0.004844658,0.002939784,0.006329879],"category_scores_gemma":[0.07879414,0.0009249591,0.002346706,0.003669685,0.005021124,0.01177568,0.009271635,0.01040834,0.001690835],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003754955,"about_ca_system_score_gemma":0.002472901,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002498483,"about_ca_topic_score_gemma":0.002355333,"domain_scores_codex":[0.9910468,0.002804746,0.0004281846,0.001633353,0.003152316,0.0009346117],"domain_scores_gemma":[0.9456099,0.03851089,0.002700408,0.008204779,0.003590686,0.001383262],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0005057506,0.0002982382,0.006476837,0.0004801776,0.0002655416,0.0002463606,0.0003960725,0.3760454,0.006569099,0.4068957,0.0131836,0.1886372],"study_design_scores_gemma":[0.00003383384,0.00009345866,0.0007701382,0.00008449538,0.00003692291,0.000140242,0.00003662942,0.7729782,0.002478186,0.2209813,0.002320517,0.00004611769],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03309784,0.00326089,0.9503787,0.001660981,0.0002243693,0.0001425103,0.0003277682,0.0009876057,0.009919372],"genre_scores_gemma":[0.7918434,0.003520814,0.1895736,0.00273317,0.001251084,0.0005956431,0.001326987,0.001124412,0.008030836],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01548253,"threshold_uncertainty_score":0.08188045,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07267578090149533,"score_gpt":0.2081383649835599,"score_spread":0.1354625840820645,"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."}}