{"id":"W4226167551","doi":"10.1609/aaai.v36i8.20894","title":"A Multi-Agent Reinforcement Learning Approach for Efficient Client Selection in Federated Learning","year":2022,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":100,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Reinforcement learning; Computer science; Federated learning; Latency (audio); Selection (genetic algorithm); Artificial intelligence; Distributed computing; Machine learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002634578,0.0008367563,0.001256134,0.0004341682,0.0007238835,0.0009882433,0.002498651,0.001232264,0.001844708],"category_scores_gemma":[0.006823548,0.0004394672,0.0004600272,0.0004578106,0.001181036,0.001695912,0.001923879,0.001833151,0.0003976899],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001075607,"about_ca_system_score_gemma":0.001694243,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004027486,"about_ca_topic_score_gemma":0.003972577,"domain_scores_codex":[0.9986861,0.00051086,0.00005746424,0.0002927741,0.0002734807,0.0001792062],"domain_scores_gemma":[0.9972342,0.001320083,0.0002835136,0.0004505317,0.0004731227,0.000238536],"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":[0.000145703,0.0001596353,0.0009200682,0.00003946607,0.00004088779,0.0001389501,0.00008422843,0.9389277,0.001942901,0.01323999,0.001463426,0.04289699],"study_design_scores_gemma":[0.000007383515,0.00001439562,0.00002259121,0.000001672953,0.000001992757,0.000009188328,0.000003647402,0.9972293,0.000243271,0.002329406,0.000134951,0.000002231836],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01687954,0.000100693,0.9806743,0.0002467482,0.00003074355,0.00005752235,0.0000253424,0.0007821748,0.001202823],"genre_scores_gemma":[0.8550471,0.00006702258,0.1423146,0.0002168831,0.00003270079,0.000142557,0.00006018812,0.00007400336,0.002044947],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004027486,"threshold_uncertainty_score":0.01393318,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09936742032616244,"score_gpt":0.3093439457427667,"score_spread":0.2099765254166043,"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."}}