{"id":"W4289317917","doi":"10.21203/rs.3.rs-1890894/v1","title":"Efficiency-optimized Data Perturbation in Local Differentially Private Federated Learning","year":2022,"lang":"en","type":"preprint","venue":"Research Square","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"Fundamental Research Funds for the Central Universities; Natural Science Foundation of Liaoning Province; National Natural Science Foundation of China","keywords":"Computer science; Differential privacy; MNIST database; Adaptability; Artificial intelligence; Robustness (evolution); Machine learning; Distributed computing; Data mining; Deep learning","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":[0.003873273,0.0006799809,0.001600313,0.0004914325,0.000897501,0.00136097,0.002515931,0.001296272,0.001323891],"category_scores_gemma":[0.009794808,0.0004284357,0.0007162651,0.001139917,0.0015134,0.003490254,0.003111979,0.001958871,0.0002973917],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001360275,"about_ca_system_score_gemma":0.001621081,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001345158,"about_ca_topic_score_gemma":0.001000897,"domain_scores_codex":[0.9960867,0.001647266,0.0001967425,0.0008855732,0.0007880306,0.0003956559],"domain_scores_gemma":[0.9952056,0.0019583,0.000341881,0.001829863,0.0004790371,0.0001852033],"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":[0.001011961,0.0002670735,0.002277725,0.000121592,0.0001091596,0.00027349,0.0002487786,0.7762057,0.007318277,0.04692717,0.004204515,0.1610346],"study_design_scores_gemma":[0.00002457663,0.0000422331,0.0001110846,0.000004941325,0.000008123965,0.00005285117,0.00001948316,0.9747297,0.002168787,0.02248635,0.000343334,0.000008569447],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0448369,0.000260923,0.9523397,0.0004289674,0.00003813838,0.00005115578,0.00009573515,0.0009682832,0.0009803069],"genre_scores_gemma":[0.9374838,0.00009474802,0.0607,0.0002379958,0.00003226874,0.0000778195,0.0001314053,0.00005873721,0.001183147],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003873273,"threshold_uncertainty_score":0.02048409,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1053796735569386,"score_gpt":0.3840149456539813,"score_spread":0.2786352720970426,"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."}}