{"type":"ScholarlyArticle","title":"Phi-Actor-Critic: Steering General-Sum Games to Pareto-Efficient Correlated Equilibria","authors":[{"name":"Wongyu Lee","slug":"wongyu-lee","given":"Wongyu","family":"Lee","affiliation":""},{"name":"Francesco Lelli","slug":"francesco-lelli","given":"Francesco","family":"Lelli","affiliation":""},{"name":"Omran Ayoub","slug":"omran-ayoub","given":"Omran","family":"Ayoub","affiliation":""},{"name":"Massimo Tornatore","slug":"massimo-tornatore","given":"Massimo","family":"Tornatore","affiliation":""}],"abstract":"Real-world multi-agent systems, from traffic coordination to resource allocation, are often modeled as general-sum games where individual incentives conflict with collective welfare. In these settings, the central challenge is not merely finding an equilibrium, but selecting socially desirable outcomes among many suboptimal Nash equilibria. \r\nStandard deep multi-agent reinforcement learning (MARL) methods struggle with this problem, as value-decomposition approaches are constrained by monotonicity assumptions and policy-gradient methods often converge to stable but socially inefficient equilibria.\r\nTo address this limitation, we propose Phi-Actor-Critic (Phi-AC), a framework that leverages swap regret minimization to steer learning toward high-welfare correlated equilibria (CE). To make counterfactual regret estimation tractable in deep MARL, Phi-AC employs a centralized attention critic that predicts vector-valued regrets in a single forward pass, avoiding computationally expensive counterfactual simulations. We further introduce a Lagrangian-based equilibrium selection mechanism that optimizes social welfare while enforcing stability through regret constraints.\r\nExperiments on matrix games, Multi-Agent Particle Environments (MPE), and the Melting Pot Harvest scenario demonstrate that Phi-AC learns efficient and stable coordination strategies across diverse mixed-motive settings while maintaining high collective return and competitive fairness.","doi":"10.24963/ijcai.2026/22","url":"https://ijcai.org/proceedings/2026/22","pdf":"https://ijcai.org/proceedings/2026/0022.pdf","markdown":"https://ijcai.org/proceedings/2026/22.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/22","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Agent-based and Multi-agent Systems","track":"Main Track","number":22,"pages":{"first":191,"last":198},"keywords":["Agent-based and Multi-agent Systems: Agent theories and models","Game Theory and Economic Paradigms: Mechanism design","Game Theory and Economic Paradigms: Computational social choice","Agent-based and Multi-agent Systems: Agent communication"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}