{"id":"W4407395225","doi":"10.1101/2025.02.10.25322018","title":"FedWeight: Mitigating Covariate Shift of Federated Learning on Electronic Health Records Data through Patients Re-weighting","year":2025,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary; University of British Columbia; McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"Natural Sciences and Engineering Research Council of Canada; Fonds de recherche du Québec – Nature et technologies; National University Health System","keywords":"Covariate; Health records; Weighting; Electronic health record; Computer science; Meaningful use; Data mining; Data science; Medicine; Machine learning; Political science; Health care","routes":{"ca_aff":true,"ca_fund":true,"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":[0.01297347,0.001098076,0.001497449,0.001184982,0.0008263178,0.001749611,0.002468188,0.001521474,0.001163741],"category_scores_gemma":[0.03489834,0.000615196,0.00145865,0.001937924,0.001175058,0.003778526,0.004179745,0.002346059,0.0008519483],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008947477,"about_ca_system_score_gemma":0.001785437,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002293519,"about_ca_topic_score_gemma":0.003050018,"domain_scores_codex":[0.9927375,0.003869633,0.0004299082,0.001841202,0.0008453006,0.0002764173],"domain_scores_gemma":[0.9814333,0.007940854,0.001439809,0.007581578,0.001209443,0.0003950539],"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":[0.00127643,0.0006662242,0.04717516,0.0003256201,0.0007037075,0.0003723191,0.001274813,0.2670599,0.009895241,0.02068055,0.01408059,0.6364894],"study_design_scores_gemma":[0.00009509818,0.0001969618,0.003616596,0.00004558407,0.00007796466,0.0002210009,0.0001333257,0.9184329,0.006278195,0.06597757,0.004878717,0.00004604712],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0398538,0.0004462697,0.9545291,0.000868678,0.00008732609,0.0001260761,0.0009099414,0.002682532,0.0004962292],"genre_scores_gemma":[0.6863483,0.0003624959,0.3052374,0.0008881437,0.0002802367,0.0003324641,0.003790662,0.0003708423,0.002389518],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01297347,"threshold_uncertainty_score":0.06861109,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0652809991062509,"score_gpt":0.3272684932587142,"score_spread":0.2619874941524633,"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."}}