{"id":"W3047438757","doi":"10.1109/infocom41043.2020.9155250","title":"Eagle: Refining Congestion Control by Learning from the Experts","year":2020,"lang":"en","type":"article","venue":"","topic":"Network Traffic and Congestion Control","field":"Computer Science","cited_by":61,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Eagle; Computer science; Reinforcement learning; Network congestion; Heuristics; Control (management); Artificial intelligence; Heuristic; Margin (machine learning); Machine learning; Artificial neural network; Network packet; Computer security","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":[],"consensus_categories":[],"category_scores_codex":[0.0001657978,0.0001256367,0.0001507605,0.000007573987,0.0002182939,0.0002198452,0.0006087819,0.00005669396,0.0001724122],"category_scores_gemma":[0.0001278074,0.00008546065,0.00005194831,0.0001387297,0.00003456543,0.0002406405,0.00007535821,0.0002140218,0.0002174466],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001181195,"about_ca_system_score_gemma":0.00003157954,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005213022,"about_ca_topic_score_gemma":0.000005415443,"domain_scores_codex":[0.9988257,0.0001891478,0.0001905325,0.0003556834,0.0002210493,0.000217888],"domain_scores_gemma":[0.9988759,0.0006439402,0.00007814926,0.0002239415,0.0000502158,0.000127836],"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":[0.00003611177,0.00001831682,0.002032435,0.000001014068,0.00005463268,0.000009684608,0.001387596,0.005804662,0.002215835,0.01668905,0.1024833,0.8692673],"study_design_scores_gemma":[0.001055565,0.0000898659,0.0004642415,0.000009168048,0.00001139271,0.000001729565,0.0001607705,0.7795246,0.00008495726,0.00006327065,0.2183656,0.0001688895],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006505273,0.0009810993,0.9286713,0.05992337,0.000260318,0.00009344923,0.000001690176,0.0005819199,0.002981608],"genre_scores_gemma":[0.9798626,0.00001948145,0.001339688,0.01777545,0.0003056506,0.00002161782,0.000006016147,0.000008143093,0.0006613343],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9733573,"threshold_uncertainty_score":0.3484984,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01538462053915904,"score_gpt":0.1912059353069693,"score_spread":0.1758213147678103,"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."}}