{"id":"W4289334177","doi":"10.21203/rs.3.rs-1891162/v1","title":"Restrictively Self-sampled and Compressed Local Differential Privacy in Federated Learning","year":2022,"lang":"en","type":"preprint","venue":"Research Square","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":3,"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; MNIST database; Differential privacy; Randomness; Sampling (signal processing); Artificial intelligence; Machine learning; Convergence (economics); Federated learning; Process (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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","metaepi_narrow","scholarly_communication","open_science","research_integrity"],"consensus_categories":["open_science"],"category_scores_codex":[0.002378658,0.0004702312,0.0006742484,0.00155819,0.0008973156,0.00136292,0.02964596,0.0005705337,0.0001208432],"category_scores_gemma":[0.02467101,0.0004975687,0.00009813841,0.001943809,0.0003608722,0.0004476896,0.45524,0.008353794,0.00001703133],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001053308,"about_ca_system_score_gemma":0.0007853241,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001407312,"about_ca_topic_score_gemma":0.00006732209,"domain_scores_codex":[0.9910686,0.002757764,0.0006120538,0.002117723,0.002097376,0.001346482],"domain_scores_gemma":[0.9901931,0.001900168,0.0002278463,0.007081618,0.000375658,0.0002216507],"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":[0.001917782,0.008379735,0.2611909,0.01398005,0.001951846,0.008708178,0.01439682,0.0186303,0.006874131,0.03597942,0.2571415,0.3708493],"study_design_scores_gemma":[0.001253055,0.0003788295,0.05790756,0.000443236,0.000007267068,0.00001310278,0.0004548188,0.8593283,0.0004996096,0.07404529,0.004978175,0.0006907077],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6024123,0.001490451,0.3722139,0.01486639,0.0006504387,0.003327628,0.0001673843,0.003758044,0.0011134],"genre_scores_gemma":[0.9663533,0.0008367159,0.03189873,0.00001388054,0.00005777505,0.0005047329,0.0002257583,0.00005436546,0.00005474552],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8406981,"threshold_uncertainty_score":0.9997476,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07673680787932541,"score_gpt":0.3684776697452106,"score_spread":0.2917408618658852,"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."}}