{"id":"W4315629593","doi":"10.1109/globecom48099.2022.10001721","title":"FedRL: Improving the Performance of Federated Learning with Non-IID Data","year":2022,"lang":"en","type":"article","venue":"GLOBECOM 2022 - 2022 IEEE Global Communications Conference","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Huawei Technologies (Canada); University of Toronto","funders":"","keywords":"Computer science; Reinforcement learning; Federated learning; Convergence (economics); Machine learning; Artificial intelligence; Data aggregator; Homogeneous; Training set; Data mining; Computer network","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":["metaepi_narrow","sts","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.001644791,0.0003058157,0.000348813,0.0001078449,0.00278822,0.0003961308,0.1044001,0.00007324788,0.0001239423],"category_scores_gemma":[0.001772097,0.0002659637,0.00005273548,0.002511168,0.0005461821,0.001115942,0.2386297,0.001512298,0.00002276637],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003456175,"about_ca_system_score_gemma":0.0007853759,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001060717,"about_ca_topic_score_gemma":0.0004302639,"domain_scores_codex":[0.9964409,0.0007071374,0.00057945,0.0008561542,0.000864725,0.000551566],"domain_scores_gemma":[0.9700099,0.0003424768,0.0006007109,0.02869168,0.000278709,0.00007648415],"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":[0.0003881791,0.00206042,0.07831537,0.0003580835,0.000985521,0.00006825113,0.001535785,0.008883523,0.0115355,0.05127713,0.3595131,0.4850791],"study_design_scores_gemma":[0.0004232819,0.0003439964,0.002218545,0.00003746096,0.00003180125,0.00009453435,0.0008770845,0.981546,0.0004014688,0.001796796,0.0118453,0.0003836763],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4875981,0.0041611,0.3954271,0.08178067,0.001940763,0.003091189,0.003108521,0.003237304,0.01965528],"genre_scores_gemma":[0.9418988,0.0004573224,0.05673428,0.0002100096,0.00001195096,0.0001883171,0.0003971671,0.00001624888,0.00008591513],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9726625,"threshold_uncertainty_score":0.9999793,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05647577360395995,"score_gpt":0.2909231474283444,"score_spread":0.2344473738243845,"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."}}