{"id":"W4388130980","doi":"10.3934/era.2023356","title":"Privacy amplification for wireless federated learning with Rényi differential privacy and subsampling","year":2023,"lang":"en","type":"article","venue":"Electronic Research Archive","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Federated learning; Differential privacy; Overhead (engineering); Computer science; MNIST database; Scheme (mathematics); Wireless; Transmission (telecommunications); Convergence (economics); Information privacy; Key (lock); Process (computing); Machine learning; Artificial intelligence; Computer network; Computer security; Telecommunications; Data mining; Deep learning; Mathematics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.001649812,0.0002314407,0.0002545906,0.0006475689,0.001160742,0.0007393149,0.01016355,0.00009479826,0.000004497367],"category_scores_gemma":[0.01062287,0.0002023645,0.00004375892,0.001557344,0.0003156278,0.0005734392,0.02564026,0.001349748,0.00003289569],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002024881,"about_ca_system_score_gemma":0.0004916452,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006273242,"about_ca_topic_score_gemma":0.00004473022,"domain_scores_codex":[0.9957231,0.0003770158,0.0002719238,0.001095113,0.0007496577,0.001783166],"domain_scores_gemma":[0.9946881,0.001859474,0.0001082003,0.002966863,0.0002285996,0.0001487498],"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":[0.001121271,0.0003665865,0.006947231,0.0005428513,0.0005563429,0.00007098099,0.00430363,0.0002105249,0.2476415,0.2513713,0.06605073,0.4208171],"study_design_scores_gemma":[0.0009528187,0.0008838423,0.004253416,0.00008514829,0.000005811078,0.00001924167,0.0001225419,0.5613568,0.01102926,0.4153802,0.005585462,0.0003254877],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3542399,0.00007728367,0.6356543,0.008105478,0.00003151283,0.00082892,0.000009868541,0.0009750886,0.00007767965],"genre_scores_gemma":[0.9625085,0.0005355755,0.03603311,0.00001224039,0.00006729209,0.0004528373,0.0001972348,0.00004544031,0.0001477549],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6082687,"threshold_uncertainty_score":0.9977111,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05946669298187045,"score_gpt":0.3439481834189567,"score_spread":0.2844814904370862,"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."}}