{"id":"W4320011078","doi":"10.1109/iccspa55860.2022.10019204","title":"DSFL: Dynamic Sparsification for Federated Learning","year":2022,"lang":"en","type":"article","venue":"","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Computer science; Bottleneck; Information bottleneck method; Key (lock); Similarity (geometry); Kernel (algebra); Theoretical computer science; Federated learning; Artificial intelligence; Machine learning; Distributed computing; Image (mathematics); Cluster analysis; Mathematics","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":["open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.000406347,0.00006972359,0.00007256238,0.00009519421,0.0005715624,0.0001365142,0.01288558,0.00002846298,0.00005564008],"category_scores_gemma":[0.00434958,0.00007353172,0.00002669862,0.0004138455,0.00001856046,0.0002843186,0.04468898,0.0002042975,0.00001954604],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000123486,"about_ca_system_score_gemma":0.00003860081,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001507621,"about_ca_topic_score_gemma":0.000005496163,"domain_scores_codex":[0.9990716,0.00005281718,0.0001292197,0.0003636275,0.0001752254,0.0002074962],"domain_scores_gemma":[0.9972782,0.000121004,0.0000689454,0.002473604,0.00003882787,0.00001947832],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002400426,0.0001641104,0.00075833,0.00002809066,0.00004388981,0.000009013178,0.0001604567,0.002340119,0.01494354,0.06740207,0.5964621,0.3176643],"study_design_scores_gemma":[0.0001361965,0.00007213973,0.0001952372,0.000001154877,0.000001259309,0.000006457597,0.00007330366,0.8981575,0.001126546,0.07046974,0.02966245,0.00009805588],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006178509,0.00003703381,0.9619582,0.02849501,0.0002325383,0.0002139698,0.000005963292,0.001724841,0.001153981],"genre_scores_gemma":[0.666767,0.000005209991,0.3320169,0.0001608716,0.000004892132,0.0001445083,0.00005113914,0.000007811112,0.0008417158],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8958173,"threshold_uncertainty_score":0.9924552,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03524959394714611,"score_gpt":0.282901786440278,"score_spread":0.2476521924931319,"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."}}