{"id":"W4286910437","doi":"10.48550/arxiv.2110.02355","title":"Robustness and sample complexity of model-based MARL for general-sum Markov games","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Ministère de la Défense Nationale","keywords":"Markov chain; Mathematics; Markov kernel; Markov process; Markov perfect equilibrium; Discrete mathematics; Markov model; Applied mathematics; Combinatorics; Mathematical optimization; Variable-order Markov model; Nash equilibrium; Statistics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01258505,0.001824504,0.002942761,0.001305011,0.001129975,0.00268017,0.003206762,0.002440688,0.003410401],"category_scores_gemma":[0.0671374,0.001114909,0.00139483,0.0006122923,0.004089877,0.004634951,0.00456208,0.005294685,0.0005204498],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004230061,"about_ca_system_score_gemma":0.004282574,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004909683,"about_ca_topic_score_gemma":0.004058528,"domain_scores_codex":[0.9942678,0.002698734,0.0002648333,0.001176576,0.000893609,0.0006984334],"domain_scores_gemma":[0.9055656,0.08215713,0.004534015,0.003763072,0.002220365,0.001759843],"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":[0.0003449943,0.0001120677,0.001283869,0.0001541076,0.00008657657,0.00006189673,0.0000839867,0.9376976,0.0008320217,0.04982494,0.0007328832,0.008785136],"study_design_scores_gemma":[0.00001439606,0.00003616106,0.00008846033,0.00001541431,0.00000617871,0.000009622026,0.000008522303,0.9810072,0.0002185905,0.01851027,0.00007754516,0.000007778553],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1022362,0.0009359966,0.8854781,0.002301944,0.00009099054,0.0002326895,0.0002893168,0.0006777114,0.007757117],"genre_scores_gemma":[0.9351048,0.0005547055,0.05981312,0.0005982526,0.0001002559,0.0004237617,0.0003892748,0.0002162419,0.002799637],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01258505,"threshold_uncertainty_score":0.06655687,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.12315007179665,"score_gpt":0.2188418402903258,"score_spread":0.0956917684936758,"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."}}