{"id":"W4225264920","doi":"10.1016/j.ins.2022.04.052","title":"CORK: A privacy-preserving and lossless federated learning scheme for deep neural network","year":2022,"lang":"en","type":"article","venue":"Information Sciences","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":37,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of New Brunswick","funders":"Ministry of Education of the People's Republic of China; Shanxi Provincial Key Research and Development Project; Natural Science Foundation of Shaanxi Province; National Natural Science Foundation of China","keywords":"Computer science; MNIST database; Artificial intelligence; Lossless compression; Counterfeit; Inference; Artificial neural network; Machine learning; Scheme (mathematics); Deep learning; Overhead (engineering); Data mining; Data compression","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":["sts","scholarly_communication","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.001859225,0.0001326091,0.0001446937,0.0002096063,0.002707256,0.001335075,0.01413651,0.00004465072,0.00002364546],"category_scores_gemma":[0.007974141,0.0001270837,0.00003206527,0.001400051,0.0001967808,0.005477718,0.06299956,0.0002940262,0.000006578374],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005935142,"about_ca_system_score_gemma":0.00008631848,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003000458,"about_ca_topic_score_gemma":0.000005280205,"domain_scores_codex":[0.9981709,0.00008628408,0.0003720939,0.0003076773,0.000578357,0.0004846638],"domain_scores_gemma":[0.9979494,0.0003488279,0.0003004592,0.001232632,0.0001095202,0.00005914042],"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":[0.00007677052,0.00006724232,0.02907881,0.0001827231,0.0000567093,0.000008307354,0.00546503,0.1606328,0.000398525,0.03951451,0.2067636,0.557755],"study_design_scores_gemma":[0.0002245309,0.0001227429,0.0007078123,0.000007624678,0.000001468604,0.00002537962,0.0006327229,0.9547137,0.00007623004,0.0253255,0.01799661,0.0001656961],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1410573,0.0002448895,0.8364473,0.01847589,0.0007332465,0.0005477308,0.00001098731,0.001023407,0.001459238],"genre_scores_gemma":[0.7839642,0.00001111508,0.2151349,0.0006932063,0.00003063613,0.0001251986,0.00001990014,0.000004233779,0.00001665343],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7940809,"threshold_uncertainty_score":0.9997016,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03948904073060502,"score_gpt":0.2869799780482666,"score_spread":0.2474909373176616,"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."}}