{"id":"W4319866023","doi":"10.1109/tdsc.2023.3241057","title":"Heterogeneous Differential-Private Federated Learning: Trading Privacy for Utility Truthfully","year":2023,"lang":"en","type":"article","venue":"IEEE Transactions on Dependable and Secure Computing","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":41,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"National Natural Science Foundation of China","keywords":"Differential privacy; Incentive; Computer science; Private information retrieval; Information privacy; Incentive compatibility; Computer security; Internet privacy; Microeconomics; Data mining; Economics","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","sts"],"consensus_categories":[],"category_scores_codex":[0.0004778723,0.0003378197,0.0003747157,0.000339206,0.00137178,0.0005411601,0.004474504,0.0002202933,0.00001837893],"category_scores_gemma":[0.0005776525,0.0003412971,0.0001513734,0.0008489183,0.00007397835,0.0004054121,0.0007326482,0.0006310852,0.00001859016],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005876329,"about_ca_system_score_gemma":0.00004732384,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001737611,"about_ca_topic_score_gemma":0.000009540379,"domain_scores_codex":[0.9974169,0.0001299875,0.0004372939,0.0009694956,0.0002986876,0.000747625],"domain_scores_gemma":[0.99719,0.000626061,0.0001432241,0.00182546,0.00007847245,0.0001367827],"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":[0.0002379206,0.0005524864,0.0003646056,0.000617045,0.0006133043,0.0002351588,0.001778119,0.02678271,0.01899735,0.001121035,0.01051392,0.9381863],"study_design_scores_gemma":[0.0006594879,0.0001796445,0.00008812105,0.00006704721,0.0000212859,0.00005516885,0.0000391607,0.9480657,0.04063316,0.008799338,0.001048853,0.0003430382],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2940257,0.00003508998,0.7018098,0.00123184,0.0005828638,0.0003177435,0.00002171902,0.001940311,0.00003486053],"genre_scores_gemma":[0.9697984,0.00005109303,0.02991431,0.0000537228,0.00004170784,0.00002848355,0.00001728783,0.00003693268,0.0000581009],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9378433,"threshold_uncertainty_score":0.9999283,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03841375796386289,"score_gpt":0.2757849288548905,"score_spread":0.2373711708910276,"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."}}