{"id":"W4312680444","doi":"10.1007/978-3-031-20050-2_41","title":"AdaBest: Minimizing Client Drift in Federated Learning via Adaptive Bias Estimation","year":2022,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University; Dalhousie University","funders":"CHIST-ERA; Research Nova Scotia; Natural Sciences and Engineering Research Council of Canada; Mitacs","keywords":"Computer science; Concept drift; Variance (accounting); Stability (learning theory); Bandwidth (computing); Federated learning; Norm (philosophy); Trust region; Adaptive learning; Data mining; Artificial intelligence; Algorithm; Machine learning; Computer network; Data stream mining","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":["metaepi_narrow","open_science","research_integrity"],"consensus_categories":["open_science"],"category_scores_codex":[0.002086661,0.0006964841,0.0007060345,0.00192723,0.0006441543,0.0007755744,0.02507181,0.0004549342,0.00005065962],"category_scores_gemma":[0.006209073,0.0007258266,0.0001153145,0.001923591,0.0006415802,0.001474351,0.07378165,0.002889742,0.00003732755],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001198872,"about_ca_system_score_gemma":0.0006385561,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001784167,"about_ca_topic_score_gemma":0.0001865608,"domain_scores_codex":[0.9939325,0.0001742064,0.0008876494,0.002456259,0.001536909,0.001012431],"domain_scores_gemma":[0.9930363,0.001193659,0.0006150124,0.004847618,0.0001781845,0.0001292825],"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":[0.00001392044,0.00005145778,0.0002247248,0.00002853082,0.00001309807,0.0003008121,0.0005000032,0.2172945,0.00009042934,0.0040217,0.0001765424,0.7772843],"study_design_scores_gemma":[0.0002465317,0.0002448777,0.0001499641,0.0003252245,0.000004324859,0.00006230631,9.371414e-7,0.819602,0.0005795222,0.1774683,0.0006893037,0.0006266781],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0004565338,0.000382201,0.9912629,0.003564767,0.001465471,0.0005668969,0.000006688164,0.0007481411,0.001546454],"genre_scores_gemma":[0.2389825,0.00006052769,0.760237,0.000476754,0.0000791472,0.00003671691,0.00002849148,0.00004785306,0.00005095697],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7766576,"threshold_uncertainty_score":0.9995193,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04055409203057733,"score_gpt":0.2657206864019703,"score_spread":0.225166594371393,"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."}}