{"id":"W4415517332","doi":"10.1016/j.compeleceng.2025.110743","title":"A comprehensive evaluation of metrics on their ability to capture the degrees of Non-IIDness with label skew in federated learning","year":2025,"lang":"en","type":"article","venue":"Computers & Electrical Engineering","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"Carleton University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Skew; Metric (unit); Overhead (engineering); Rank (graph theory); Performance metric; Computation; Learning to rank; Strengths and weaknesses","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007104981,0.0001865684,0.0003411948,0.0005710049,0.00005570968,0.00005205765,0.004904598,0.00009845358,3.803069e-7],"category_scores_gemma":[0.005773711,0.0001335447,0.00003572895,0.004689538,0.00003770886,0.0001186677,0.004467341,0.0005133031,4.885418e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002324011,"about_ca_system_score_gemma":0.0001381199,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006539518,"about_ca_topic_score_gemma":0.000005979075,"domain_scores_codex":[0.9984202,0.0001318487,0.0003139661,0.0003972157,0.0004464649,0.0002903398],"domain_scores_gemma":[0.9967349,0.001172463,0.00009511559,0.001595492,0.000365869,0.00003616495],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004031629,0.0001540538,0.001040521,0.000115621,0.00007207086,0.000003827715,0.0002829247,0.8444002,0.00743264,0.001542245,0.0008312456,0.1440843],"study_design_scores_gemma":[0.0004703089,0.0001869827,0.01577093,0.0001763157,0.00001019887,0.000001807324,0.00001363209,0.9769174,0.005762833,0.0005235153,0.00004385816,0.0001221456],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3091965,0.0002291407,0.6889442,0.00102723,0.00009669548,0.0003490504,0.000001147545,0.0001282255,0.00002778728],"genre_scores_gemma":[0.9063033,0.000007305614,0.09359263,0.00005523691,0.000005470592,0.00002501438,0.000002308504,0.000007683034,9.898708e-7],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5971068,"threshold_uncertainty_score":0.9114045,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02739944413914019,"score_gpt":0.275127713051404,"score_spread":0.2477282689122638,"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."}}