{"id":"W3181858657","doi":"10.1007/978-3-030-81242-3_2","title":"Comparing Local and Central Differential Privacy Using Membership Inference Attacks","year":2021,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Differential privacy; Inference; Computer science; Upper and lower bounds; Differential (mechanical device); Data mining; Artificial intelligence; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.0004588354,0.000776881,0.0008906384,0.0007463905,0.0003858356,0.001383642,0.02770558,0.0005842975,0.00004223197],"category_scores_gemma":[0.002895871,0.0007714266,0.0001236934,0.0007401758,0.001905283,0.001160555,0.1236005,0.001719028,0.000007352789],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006180809,"about_ca_system_score_gemma":0.0007347477,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007948608,"about_ca_topic_score_gemma":0.0001128005,"domain_scores_codex":[0.9942487,0.0000728871,0.000688864,0.00249696,0.001198513,0.001294051],"domain_scores_gemma":[0.9908502,0.000684249,0.0003634388,0.007602694,0.0002159596,0.0002834505],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002996078,0.0001649862,0.006659855,0.0004878124,0.0001331916,0.001072247,0.001348139,0.02863406,0.001741652,0.05049194,0.000361371,0.9088748],"study_design_scores_gemma":[0.0002685608,0.00005132664,0.0006591016,0.000704519,0.00001421428,0.0001331579,4.326312e-7,0.8158479,0.002007128,0.1794116,0.0001617529,0.00074027],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004379012,0.0006565547,0.9903179,0.001539803,0.001792621,0.0003074958,0.000008051605,0.0004396388,0.000558897],"genre_scores_gemma":[0.5823761,0.00006927672,0.4171277,0.0002096938,0.0001560325,0.000002866919,0.000009309335,0.00002907153,0.00001990552],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9081345,"threshold_uncertainty_score":0.999653,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06331319455675687,"score_gpt":0.2916578776809519,"score_spread":0.228344683124195,"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."}}