{"id":"W3037719421","doi":"10.65109/gjmw6851","title":"Neural Replicator Dynamics: Multiagent Learning via Hedging Policy Gradients","year":2020,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Softmax function; Reinforcement learning; Computer science; Nash equilibrium; Mathematical optimization; Replicator equation; Convergence (economics); Gradient descent; Best response; Regret; Margin (machine learning); Artificial intelligence; Artificial neural network; Applied mathematics; Mathematical economics; Mathematics; Machine learning; Economics","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":[0.0001457994,0.0001882958,0.0001775875,0.0001143685,0.0002170338,0.0002261269,0.001061538,0.00005057367,0.00003102102],"category_scores_gemma":[0.0002957576,0.0001802672,0.00009342973,0.0006284575,0.00003578193,0.0004180685,0.0006984731,0.0003521763,0.0003430731],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001437006,"about_ca_system_score_gemma":0.00004501594,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008140106,"about_ca_topic_score_gemma":0.000001472901,"domain_scores_codex":[0.9982477,0.00007588849,0.0003170505,0.0005427436,0.0003506899,0.0004658558],"domain_scores_gemma":[0.9989693,0.00006185264,0.0001003842,0.000512222,0.00005426923,0.0003019467],"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":[0.000004478569,0.00001509261,0.01046987,0.00002349779,0.00002422319,0.0000218323,0.001408581,0.9374681,0.0005712582,0.02250263,0.0002616176,0.02722889],"study_design_scores_gemma":[0.0002393388,0.0001216607,0.001194177,0.000005712605,0.000003393012,0.000009051289,0.00004826671,0.9953075,0.0003114828,0.00003581058,0.002522107,0.0002015069],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008428235,0.00001265383,0.978905,0.007503328,0.0001731469,0.0001677938,2.036193e-7,0.0007656591,0.004044007],"genre_scores_gemma":[0.9661419,0.000005452227,0.02956753,0.003094923,0.0001610137,0.000007266315,0.00000647984,0.00002179634,0.0009936497],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9577137,"threshold_uncertainty_score":0.7351085,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01798073952334079,"score_gpt":0.2518673730577505,"score_spread":0.2338866335344097,"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."}}