{"id":"W2567588361","doi":"10.4230/lipics.socg.2017.45","title":"Lower Bounds for Differential Privacy from Gaussian Width","year":2017,"lang":"en","type":"article","venue":"DROPS (Schloss Dagstuhl – Leibniz Center for Informatics)","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Differential privacy; Polytope; Gaussian; Workload; Convex geometry; Sample (material); Measure (data warehouse); Mathematics; Convex body; Sensitivity (control systems); Upper and lower bounds; Regular polygon; Computer science; Discrete mathematics; Algorithm; Convex optimization; Convex set; Mathematical analysis; Database; Geometry","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":[0.01036963,0.002784092,0.003646169,0.002723681,0.001802957,0.0068701,0.005022282,0.003166324,0.006858436],"category_scores_gemma":[0.08782321,0.001430567,0.002877089,0.003334382,0.005674566,0.01887568,0.007314894,0.008514309,0.001230212],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006337948,"about_ca_system_score_gemma":0.002548257,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001127286,"about_ca_topic_score_gemma":0.0007902951,"domain_scores_codex":[0.987251,0.003647158,0.0005634896,0.002529888,0.003972051,0.002036441],"domain_scores_gemma":[0.8862796,0.09012882,0.004817726,0.01130785,0.004135367,0.003330589],"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":[0.001518886,0.0004675678,0.006607831,0.0007325742,0.0003433438,0.0003925003,0.001053009,0.2537307,0.01575445,0.6640825,0.006345233,0.04897142],"study_design_scores_gemma":[0.00005233688,0.0001939473,0.000996075,0.00007170624,0.00006907592,0.0002094018,0.0001131084,0.5668906,0.004098595,0.4255052,0.001745347,0.00005455363],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09582011,0.002356106,0.883313,0.004274686,0.0001683044,0.0002554337,0.0009942778,0.0007521284,0.01206597],"genre_scores_gemma":[0.8449674,0.002677604,0.1390226,0.001655653,0.001108717,0.001201117,0.001362797,0.0009522741,0.007051723],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01036963,"threshold_uncertainty_score":0.05484051,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02974776109486496,"score_gpt":0.2898978853087917,"score_spread":0.2601501242139267,"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."}}