{"id":"W3204293975","doi":"10.1007/s13235-023-00490-2","title":"Robustness and Sample Complexity of Model-Based MARL for General-Sum Markov Games","year":2023,"lang":"en","type":"article","venue":"Dynamic Games and Applications","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"Ministère de la Défense Nationale","keywords":"Markov chain; Mathematics; Markov kernel; Markov perfect equilibrium; Markov process; Markov model; Discrete mathematics; Applied mathematics; Mathematical optimization; Combinatorics; Mathematical economics; Variable-order Markov model; Nash equilibrium; Statistics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.01233857,0.002321487,0.00389913,0.00246063,0.001544162,0.004371957,0.004378589,0.003195235,0.004917636],"category_scores_gemma":[0.07805885,0.001798527,0.001807843,0.001166642,0.005416887,0.007201931,0.00569112,0.005689342,0.0005708925],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00489511,"about_ca_system_score_gemma":0.003953475,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005262939,"about_ca_topic_score_gemma":0.004267949,"domain_scores_codex":[0.9935184,0.003100522,0.0002807371,0.001268898,0.001095371,0.0007360735],"domain_scores_gemma":[0.8547907,0.1276272,0.006571738,0.005080716,0.003548162,0.002381394],"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":[0.0004235278,0.0001430576,0.001341943,0.0001745004,0.0001054531,0.00008807086,0.0001185654,0.8952778,0.0007886318,0.09372538,0.001185075,0.006628063],"study_design_scores_gemma":[0.00001365357,0.00002815587,0.0001216286,0.00001090587,0.000007898478,0.00001059683,0.000009593284,0.9598213,0.0001633885,0.03974333,0.00005679744,0.00001263981],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1410721,0.0009144432,0.8463246,0.001931114,0.0000937364,0.0001997653,0.0006093227,0.000711089,0.008143879],"genre_scores_gemma":[0.9538854,0.0004337186,0.03996526,0.0003687257,0.0001619135,0.0003234376,0.0006690613,0.0003300457,0.003862431],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01233857,"threshold_uncertainty_score":0.06525332,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03328730004061116,"score_gpt":0.288582032584233,"score_spread":0.2552947325436219,"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."}}