{"id":"W2163187564","doi":"10.48550/arxiv.1503.08329","title":"Risk Bounds for the Majority Vote: From a PAC-Bayesian Analysis to a Learning Algorithm","year":2015,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":91,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"","keywords":"Bayesian probability; Statistical learning theory; Computer science; Upper and lower bounds; Divergence (linguistics); AdaBoost; Artificial intelligence; Probably approximately correct learning; Machine learning; Algorithm; Bayesian statistics; Mathematics; Support vector machine; Bayesian inference; Algorithmic learning theory; Unsupervised 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":[0.01222439,0.001542801,0.002725257,0.001920375,0.001443888,0.004670459,0.003599334,0.002962837,0.006149835],"category_scores_gemma":[0.05093639,0.001138587,0.001585195,0.00198084,0.005003858,0.008803708,0.005063034,0.007071361,0.001496803],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002880581,"about_ca_system_score_gemma":0.002398715,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001492652,"about_ca_topic_score_gemma":0.0009294066,"domain_scores_codex":[0.9904795,0.004597498,0.0002743212,0.00122407,0.002836426,0.0005882059],"domain_scores_gemma":[0.9745907,0.01948186,0.001091052,0.002137594,0.002217453,0.0004814421],"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":[0.000126429,0.00007063209,0.00074621,0.0001903824,0.00008191809,0.00007104417,0.0002642788,0.222528,0.0009186438,0.6907874,0.006557565,0.07765753],"study_design_scores_gemma":[0.00001478312,0.00004858782,0.0002005704,0.00006781596,0.00001444443,0.00006113068,0.00003055443,0.5207208,0.0007923243,0.4745091,0.003512532,0.00002718033],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004089502,0.0009711621,0.9887562,0.001213816,0.00006565113,0.00003516097,0.00004670343,0.0001277697,0.004693949],"genre_scores_gemma":[0.457112,0.003357322,0.5220872,0.002024133,0.0013228,0.0007200622,0.0003580766,0.0009039359,0.01211451],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01222439,"threshold_uncertainty_score":0.06464952,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03610584429556141,"score_gpt":0.1980005763968624,"score_spread":0.161894732101301,"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."}}