{"id":"W4403517697","doi":"10.3233/faia240816","title":"Mask-Encoded Sparsification: Mitigating Biased Gradients in Communication-Efficient Split Learning","year":2024,"lang":"en","type":"book-chapter","venue":"Frontiers in artificial intelligence and applications","topic":"Machine Learning and ELM","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fundamental Research Funds for the Central Universities; Natural Science Foundation of Jiangsu Province; Government of Jiangsu Province; Canadian Institute for Advanced Research; China Postdoctoral Science Foundation; National Natural Science Foundation of China","keywords":"Computer science; Artificial intelligence","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001040941,0.0009103765,0.00071779,0.0004225997,0.0003618078,0.0008803529,0.001374517,0.001138281,0.003387127],"category_scores_gemma":[0.005382849,0.0003418323,0.0002810766,0.0005102271,0.001114575,0.002180035,0.002389398,0.00146926,0.001061517],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004201454,"about_ca_system_score_gemma":0.0007965807,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007921733,"about_ca_topic_score_gemma":0.001286226,"domain_scores_codex":[0.9996015,0.0001204641,0.00002033286,0.00006587294,0.0001464871,0.00004529614],"domain_scores_gemma":[0.9986998,0.0007390272,0.00008710215,0.0002685182,0.0001413287,0.00006414502],"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":[0.0005154865,0.0001352678,0.001069435,0.000288745,0.00004947512,0.0003022883,0.0003437862,0.4540197,0.03030078,0.06816736,0.007990071,0.4368175],"study_design_scores_gemma":[0.00001292721,0.00006039771,0.0000795685,0.0000178506,0.000005854948,0.00008697285,0.00002215166,0.9750314,0.00695295,0.01628025,0.001442081,0.000007486172],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01863476,0.0003641269,0.9778788,0.0002756294,0.00004402057,0.00004396578,0.00007058189,0.0005068238,0.002181385],"genre_scores_gemma":[0.5012668,0.0007037591,0.4900681,0.0005348407,0.0001538057,0.0002461419,0.0004426355,0.0003193322,0.006264567],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003387127,"threshold_uncertainty_score":0.01133108,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0347162747366038,"score_gpt":0.273585018793439,"score_spread":0.2388687440568353,"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."}}