{"passport":{"slug":"alternating-marl","display_name":"ALTERNATING-MARL","model_type":"null","parameter_count":null,"context_window":null,"claimed_developer":"arXiv","confirmed_developer":null,"developer_confirmed_at":null,"developer_source":null,"first_appeared_at":"2026-03-11T06:32:35.047014+00:00","first_appeared_on":"rss_pipeline","status":"confirmed","availability":null,"deployment_status":"not_deployed","availability_scope":null,"gate":1,"eu_ai_act":null,"regulatory_status":{},"open_source":null,"weights_available":false,"license":null,"provenance_chain":[],"parent_model":null,"model_family":null,"superseded_by":null,"supersedes":[],"recommended_replacement":null,"product_links":null},"events":[{"event_type":"first_appearance","event_date":"2026-03-11T06:32:35.047014+00:00","title":"First recorded in AI news","detail":"The paper proposes an alternating learning framework called ALTERNATING-MARL for cooperative Markov games with a global agent and n homogeneous local agents under communication constraints. The framework involves subsampled mean-field Q-learning by the global agent and local policy optimization, and it is shown to converge to an approximate Nash Equilibrium.","source_url":"https://arxiv.org/abs/2603.03759","source_name":"ArXiv AI"}],"delta":{"days_unattributed":121,"claimed_developer":"arXiv","confirmed_developer":null,"match":false}}