{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001441377,0.0001000328,0.0001481836,0.00008655929,0.0001376838,0.0000630897,0.0002794804,0.0000403285,0.000002110467],"category_scores_gemma":[0.00002251262,0.0000985618,0.00003707507,0.0002665231,0.0001612985,0.0000761188,0.00014385,0.00005141014,8.620482e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000135336,"about_ca_system_score_gemma":0.00003472294,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001620581,"about_ca_topic_score_gemma":0.000007849955,"domain_scores_codex":[0.999242,0.0000132282,0.0001907409,0.0002727817,0.00009978306,0.0001814353],"domain_scores_gemma":[0.9991891,0.0002142952,0.00009729708,0.0003737983,0.00006848539,0.00005701121],"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":[0.000002175447,0.00001293283,0.0003707788,0.00008779403,0.000008849775,5.936842e-8,0.00005417593,0.9163953,0.0001839275,0.06573028,0.0001718027,0.0169819],"study_design_scores_gemma":[0.0002275565,0.00002288026,0.0030519,0.000009925704,0.000008203209,6.777364e-7,0.00002319778,0.9870986,0.00002584287,0.008479628,0.0009460167,0.0001055091],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01300241,0.00003925124,0.9852847,0.0009497153,0.0000206636,0.0004450419,0.00006075907,0.0001108575,0.00008659951],"genre_scores_gemma":[0.4628824,0.00009039913,0.5356756,0.0001077352,0.00001741341,0.0003492098,0.0001192834,0.00001446199,0.0007435199],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.44988,"threshold_uncertainty_score":0.4019234,"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."}}