{"id":"W3153573172","doi":"10.1109/tnnls.2021.3070584","title":"Multiagent Meta-Reinforcement Learning for Adaptive Multipath Routing Optimization","year":2021,"lang":"en","type":"article","venue":"IEEE Transactions on Neural Networks and Learning Systems","topic":"Software-Defined Networks and 5G","field":"Computer Science","cited_by":78,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; Toronto Metropolitan University","funders":"National Natural Science Foundation of China","keywords":"Reinforcement learning; Computer science; Markov decision process; Routing (electronic design automation); Node (physics); Multipath routing; Adaptive routing; Multi-agent system; Distributed computing; Network packet; Artificial intelligence; Markov process; Dynamic Source Routing; Computer network; Routing protocol; Engineering","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"],"consensus_categories":[],"category_scores_codex":[0.000509577,0.0003104314,0.0004540009,0.00009212373,0.00110626,0.0004056722,0.0001875703,0.0001453066,0.00001420878],"category_scores_gemma":[0.000031344,0.0002743007,0.0002689543,0.0003901318,0.00002800146,0.0003363917,0.000009325323,0.0007124746,0.000002069303],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005242745,"about_ca_system_score_gemma":0.00002792408,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007830113,"about_ca_topic_score_gemma":0.000007747971,"domain_scores_codex":[0.9977415,0.0004123382,0.0004622731,0.0006353627,0.0002635146,0.0004850299],"domain_scores_gemma":[0.9984028,0.0007476597,0.0002370416,0.0002586459,0.0002018339,0.0001519877],"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":[0.00002855953,0.00003360211,0.00002808713,0.00001967959,0.0002670367,0.00001012001,0.0003184465,0.9805138,0.00001282168,0.0001572755,0.0000229824,0.01858756],"study_design_scores_gemma":[0.0006909758,0.0003332527,0.00001618454,0.00006744985,0.0002060104,0.00003728155,0.0002128866,0.9973266,0.00005642631,0.000001603046,0.0007565544,0.000294795],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0008530956,0.001281045,0.9955536,0.0001048576,0.001355334,0.0004696408,0.000001184514,0.0003448819,0.00003632984],"genre_scores_gemma":[0.9884639,0.0001908865,0.009595639,0.0001284498,0.0001939601,0.0002291774,0.00001119401,0.00003690804,0.001149887],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9876108,"threshold_uncertainty_score":0.9999709,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.038082896844584,"score_gpt":0.2436636899350177,"score_spread":0.2055807930904338,"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."}}