{"paper":{"title":"DelAC: A Multi-agent Reinforcement Learning of Team-Symmetric Stochastic Games","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Team-symmetric stochastic games always have a team-symmetric Nash equilibrium that a new actor-critic algorithm can locate efficiently.","cross_cats":["cs.GT"],"primary_cat":"cs.MA","authors_text":"Duan-Shin Lee, Yu-Hsiu Hung","submitted_at":"2026-05-11T12:00:27Z","abstract_excerpt":"In this paper we study team-symmetric games with $m\\ge 2$ teams. Players within a team have symmetric identity and have a common payoff function. We show that team-symmetric games always have a team-symmetric Nash equilibrium. We develop and solve a linear complementarity problem of team-symmetric Nash equilibria. We propose an actor-critic based multi-agent reinforcement learning algorithm for team-symmetric games. Through simulations, we show that this multi-agent reinforcement learning algorithm performs much better than many existing algorithms."},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We show that team-symmetric games always have a team-symmetric Nash equilibrium. We develop and solve a linear complementarity problem of team-symmetric Nash equilibria. ... this multi-agent reinforcement learning algorithm performs much better than many existing algorithms.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that players within a team have perfectly symmetric identities and identical payoff functions holds in the target applications, and that simulation results generalize beyond the tested environments.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Team-symmetric games always have team-symmetric Nash equilibria solvable via linear complementarity problems, and the DelAC actor-critic MARL algorithm outperforms existing methods in simulations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Team-symmetric stochastic games always have a team-symmetric Nash equilibrium that a new actor-critic algorithm can locate efficiently.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"b5e3331057af1c7ef5f956f3b5a84f3d1e9ed21ceda2e4707bdb593055deb739"},"source":{"id":"2605.12555","kind":"arxiv","version":1},"verdict":{"id":"dc81200c-9ceb-4bb6-afae-c83033e6f981","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:26:41.584151Z","strongest_claim":"We show that team-symmetric games always have a team-symmetric Nash equilibrium. We develop and solve a linear complementarity problem of team-symmetric Nash equilibria. ... this multi-agent reinforcement learning algorithm performs much better than many existing algorithms.","one_line_summary":"Team-symmetric games always have team-symmetric Nash equilibria solvable via linear complementarity problems, and the DelAC actor-critic MARL algorithm outperforms existing methods in simulations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that players within a team have perfectly symmetric identities and identical payoff functions holds in the target applications, and that simulation results generalize beyond the tested environments.","pith_extraction_headline":"Team-symmetric stochastic games always have a team-symmetric Nash equilibrium that a new actor-critic algorithm can locate efficiently."},"references":{"count":35,"sample":[{"doi":"","year":2024,"title":"S. V. Albrecht, F. Christianos, and L. Sch ¨afer,Multi-Agent Reinforcement Learning Foundations and Modern Approaches. Cambridge, Massachusetts: The MIT Press, 2024","work_id":"938b75ad-a62e-4d7b-9913-82903460cfa0","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2007,"title":"If multi-agent learning is the answer, what is the question?","work_id":"3edec534-8da1-4f61-b699-2365de22f47f","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2003,"title":"Nash Q-learning for general-sum stochastic games,","work_id":"00f596d0-9d08-4f30-8750-df4841e06a4d","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2006,"title":"The complexity of computing a Nash equilibrium,","work_id":"cbc8089f-d08d-4046-8647-9ecd91285e11","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2006,"title":"Settling the complexity of two-player Nash equilibrium,","work_id":"f3fa2a8f-9f41-43ba-a45c-c9769efc1210","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":35,"snapshot_sha256":"9cf9b0dc5e398e23238c8b1cd5de5786baace12377a2dd690aa374f6a6a1b01a","internal_anchors":3},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}