{"id":"W2962851649","doi":"","title":"Distributed Maximization of \"Submodular plus Diversity\" Functions for Multi-label Feature Selection on Huge Datasets","year":2019,"lang":"en","type":"article","venue":"International Conference on Artificial Intelligence and Statistics","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Submodular set function; Feature selection; Computer science; Redundancy (engineering); Automatic summarization; Feature (linguistics); Greedy algorithm; Maximization; Optimization problem; Mathematical optimization; Data mining; Function (biology); Artificial intelligence; Machine learning; Algorithm; Mathematics","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":[0.003814315,0.001215982,0.00217921,0.000774037,0.000797946,0.001125986,0.001835589,0.001708467,0.001961832],"category_scores_gemma":[0.00704412,0.0005119257,0.0008652164,0.00169212,0.001142588,0.002037237,0.001509801,0.001610236,0.0005263346],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001561339,"about_ca_system_score_gemma":0.001341739,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001896384,"about_ca_topic_score_gemma":0.00281232,"domain_scores_codex":[0.9982639,0.0008531419,0.00004637552,0.0003392519,0.0003309491,0.0001663355],"domain_scores_gemma":[0.9949517,0.003703769,0.0003548324,0.0003505099,0.0004721114,0.0001670928],"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":[0.0002389155,0.0001980788,0.0007658927,0.0001369978,0.00008433851,0.0001107896,0.00008399056,0.8548027,0.003137681,0.01067923,0.005122012,0.1246394],"study_design_scores_gemma":[0.00002492514,0.0000362532,0.0001333231,0.000003569556,0.000005464669,0.00001794145,0.00001070689,0.990409,0.0004217185,0.008691688,0.000240697,0.000004738061],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01337424,0.0002596065,0.9847872,0.0004095742,0.00002403398,0.00005362656,0.00007094493,0.0003028501,0.000718007],"genre_scores_gemma":[0.5705342,0.0003469801,0.423615,0.0005126413,0.0002735956,0.0004992236,0.0006399414,0.0001740553,0.003404437],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003814315,"threshold_uncertainty_score":0.02017224,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1479739312535318,"score_gpt":0.3478739285134115,"score_spread":0.1998999972598797,"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."}}