{"id":"W3047304572","doi":"10.1109/infocom41043.2020.9155494","title":"Optimizing Federated Learning on Non-IID Data with Reinforcement Learning","year":2020,"lang":"en","type":"article","venue":"","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":958,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; University of Toronto","funders":"","keywords":"Computer science; Upload; Reinforcement learning; Independent and identically distributed random variables; Artificial intelligence; Machine learning; Mobile device; Convergence (economics); Software deployment; Federated learning; Distributed computing; Data modeling; Database; World Wide Web","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":[0.0003764197,0.0002292742,0.0002166966,0.00008647676,0.0004087297,0.0005754552,0.02501673,0.00008551364,0.00006412699],"category_scores_gemma":[0.008053926,0.0001854815,0.00001966947,0.0006791684,0.00005336691,0.001249506,0.1004879,0.0008568788,0.0001983693],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005218387,"about_ca_system_score_gemma":0.00007914034,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004164914,"about_ca_topic_score_gemma":0.000004162114,"domain_scores_codex":[0.9977766,0.00006466271,0.0002554053,0.0009728666,0.0004759308,0.0004545063],"domain_scores_gemma":[0.994159,0.000143984,0.0001467235,0.005365396,0.0000606063,0.000124304],"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":[0.0002089373,0.0001012397,0.002189362,0.0001233129,0.0002848374,0.0004130322,0.001240921,0.5427329,0.004965888,0.004294407,0.3449956,0.09844957],"study_design_scores_gemma":[0.000402087,0.0005646513,0.00002725144,0.00005769151,0.000004825705,0.00000617322,0.0001619369,0.9830156,0.004860551,0.0001255052,0.01050084,0.0002728703],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001690597,0.00001772623,0.9515656,0.02983018,0.00005980058,0.0001824919,0.000001004321,0.002334536,0.01431801],"genre_scores_gemma":[0.5720934,0.00002554957,0.4264742,0.001047738,0.00003354284,0.000007124585,0.00007478353,0.000019671,0.0002238802],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5704029,"threshold_uncertainty_score":0.9802584,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05745084960832399,"score_gpt":0.2763946947713953,"score_spread":0.2189438451630714,"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."}}