{"id":"W4319069175","doi":"10.1109/tpds.2023.3241965","title":"Synchronize Only the Immature Parameters: Communication-Efficient Federated Learning By Freezing Parameters Adaptively","year":2023,"lang":"en","type":"article","venue":"IEEE Transactions on Parallel and Distributed Systems","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Chinese University of Hong Kong; National Natural Science Foundation of China","keywords":"Bottleneck; Computer science; Python (programming language); Edge device; Bandwidth (computing); Distributed computing; Enhanced Data Rates for GSM Evolution; Process (computing); Synchronization (alternating current); Overhead (engineering); Artificial intelligence; Computer network; Operating system; Embedded system","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":["sts","open_science"],"consensus_categories":[],"category_scores_codex":[0.0006823643,0.0003165752,0.0003265024,0.000152324,0.001389491,0.0008065725,0.006051566,0.0002394307,0.000002886253],"category_scores_gemma":[0.0005335068,0.0002451792,0.0001004673,0.001284689,0.0002501187,0.0003264876,0.0004906688,0.0008927977,0.00006742157],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001460648,"about_ca_system_score_gemma":0.00007497363,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004914795,"about_ca_topic_score_gemma":0.00003276525,"domain_scores_codex":[0.9974635,0.0004664431,0.0004574265,0.0006320839,0.0004339664,0.0005466023],"domain_scores_gemma":[0.9958733,0.0009912213,0.0002185094,0.002698396,0.0001027136,0.0001158268],"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":[0.00006087865,0.0001527415,0.00003290653,0.00005369563,0.0002307469,0.00001823132,0.0003149007,0.9125074,0.0006262038,0.0004057862,0.07911492,0.006481572],"study_design_scores_gemma":[0.0005069034,0.0001054417,0.0001146787,0.0001524475,0.00002546914,0.00004873504,0.001257738,0.9946965,0.0002768642,0.0005631442,0.001922898,0.0003291964],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03253489,0.0007885717,0.9566517,0.007062569,0.000357117,0.0005349112,0.0004464735,0.001579301,0.0000444504],"genre_scores_gemma":[0.9936123,0.0003759575,0.005450022,0.00006003628,0.000006344001,0.0002004043,0.0001598188,0.00002158161,0.0001135444],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9610774,"threshold_uncertainty_score":0.9999106,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02801249381097042,"score_gpt":0.2487432585051979,"score_spread":0.2207307646942274,"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."}}