{"id":"W2886444620","doi":"10.48550/arxiv.1808.04866","title":"Mitigating Sybils in Federated Learning Poisoning","year":2018,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":362,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Business; Criminology; Internet privacy; Computer science; Computer security; Psychology","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","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.0007223,0.0003917443,0.0004213537,0.0005594251,0.0003230213,0.0004055457,0.02435235,0.0005681798,0.0000262378],"category_scores_gemma":[0.008974891,0.000492681,0.00011758,0.001344915,0.0002199448,0.0007858553,0.1589297,0.001911363,0.0001098518],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004777254,"about_ca_system_score_gemma":0.0002261045,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003579971,"about_ca_topic_score_gemma":0.00009745274,"domain_scores_codex":[0.9968435,0.000297686,0.0003282539,0.001720676,0.0001345878,0.0006752687],"domain_scores_gemma":[0.9934404,0.0002790633,0.0004030522,0.005605218,0.0001649604,0.0001072778],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001520727,0.0006993925,0.4085422,0.001193857,0.0007979753,0.009720996,0.002211872,0.3862203,0.002416601,0.1067855,0.0534267,0.02783254],"study_design_scores_gemma":[0.0002921644,0.00004067311,0.001214819,0.0004360471,0.00001084414,0.000006142966,0.000120828,0.8267199,0.0007193408,0.1698245,0.0001354683,0.0004791699],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5996075,0.00005930986,0.3944606,0.0008595562,0.0004597057,0.0002303846,0.000005658649,0.001479222,0.002838101],"genre_scores_gemma":[0.9588453,0.00008060251,0.04070448,0.00005535816,0.00005102191,0.000001528997,0.00002346253,0.00002602216,0.0002122367],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4404996,"threshold_uncertainty_score":0.9997525,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07659239540079592,"score_gpt":0.2143127364843953,"score_spread":0.1377203410835994,"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."}}