Cooperative Game-theoretic Credit Assignment For Multi-agent Policy Gradients Via The Core

Abstract

This work focuses on the credit assignment problem in cooperative multi-agent reinforcement learning (MARL). Sharing the global advantage among agents often leads to insufficient policy optimization, as it fails to capture the coalitional contributions of different agents. In this work, we revisit the policy update process from a coalitional perspective and propose CORA, an advantage allocation method guided by a cooperative game-theoretic core allocation. By evaluating the marginal contributions of different coalitions and combining clipped double Q-learning to mitigate overestimation bias, CORA estimates coalition-wise advantages. The core formulation enforces coalition-wise lower bounds on allocated credits, so that coalitions with higher advantages receive stronger total incentives for their participating agents, enabling the global advantage to be attributed to different coalition strategies and promoting coordinated optimal behavior. To reduce computational overhead, we employ ra

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyji2025cooperative

Related papers

• Learning Explicit Credit Assignment For Cooperative Multi-agent Reinforcement Learning Via Polarization Policy Gradient (2022)4.52
• Nucleolus Credit Assignment For Effective Coalitions In Multi-agent Reinforcement Learning (2025)0.00
• Cooperative Multi-agent Transfer Learning With Level-adaptive Credit Assignment (2021)0.00
• Learning Implicit Credit Assignment For Cooperative Multi-agent Reinforcement Learning (2020)0.00
• Adaptive Value Decomposition With Greedy Marginal Contribution Computation For Cooperative Multi-agent Reinforcement Learning (2023)3.58
• Counterfactual Multi-agent Policy Gradients (2017)0.00
• Asynchronous Credit Assignment For Multi-agent Reinforcement Learning (2024)0.00
• Assigning Credit With Partial Reward Decoupling In Multi-agent Proximal Policy Optimization (2024)0.00