{"id":"W3091802565","doi":"10.1145/3417994","title":"Near-optimal Sample Complexity Bounds for Robust Learning of Gaussian Mixtures via Compression Schemes","year":2020,"lang":"en","type":"article","venue":"Journal of the ACM","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; University of British Columbia; University of Waterloo; McMaster University; Vector Institute","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Mixture model; Upper and lower bounds; Compression (physics); Gaussian; Sample complexity; Matching (statistics); Mathematics; Scheme (mathematics); Data compression; Algorithm; Class (philosophy); Computer science; Distribution (mathematics); Artificial intelligence; Statistics; Mathematical analysis; Physics","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.0003982081,0.0001050181,0.0002566153,0.00003174081,0.0002876024,0.0001040159,0.002435141,0.00004094409,0.0000224218],"category_scores_gemma":[0.001464734,0.0000652084,0.0002186795,0.000187647,0.00009093715,0.0001896142,0.0007811951,0.0004794477,0.000001116277],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001380263,"about_ca_system_score_gemma":0.00006605247,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003023809,"about_ca_topic_score_gemma":6.210116e-7,"domain_scores_codex":[0.9988822,0.0001475929,0.0003416935,0.0001297352,0.0003306041,0.0001681729],"domain_scores_gemma":[0.9984907,0.0002498562,0.0006028445,0.000420891,0.0001316914,0.0001040353],"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":[0.001188957,0.0008156718,0.09101736,0.000747303,0.0007767118,0.00005179872,0.027628,0.5478532,0.04929101,0.009384706,0.04238997,0.2288553],"study_design_scores_gemma":[0.0006706013,0.0005040911,0.007741818,0.00009389814,0.00002349414,0.00004594838,0.00006091713,0.9677072,0.005059214,0.003039816,0.01493329,0.0001197735],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07360693,0.0001581728,0.9145641,0.01124015,0.0002911483,0.00006783254,0.000002745277,0.00002361934,0.00004529602],"genre_scores_gemma":[0.4628064,0.000002647744,0.5367565,0.0002210446,0.0001815138,4.440623e-7,5.580909e-7,0.000006517113,0.00002435902],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4198539,"threshold_uncertainty_score":0.4525138,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04636202812958371,"score_gpt":0.2856437907121075,"score_spread":0.2392817625825238,"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."}}