Efficient Reinforcement Learning for Zero-Shot Coordination in Evolving Games
This addresses a computational bottleneck for researchers in multi-agent reinforcement learning, though it is incremental as it builds on existing population-based methods.
The paper tackles the computational inefficiency of scaling population size in zero-shot coordination for reinforcement learning by proposing Scalable Population Training (ScaPT), which achieves superior performance in the Hanabi cooperative game.
Zero-shot coordination(ZSC), a key challenge in multi-agent game theory, has become a hot topic in reinforcement learning (RL) research recently, especially in complex evolving games. It focuses on the generalization ability of agents, requiring them to coordinate well with collaborators from a diverse, potentially evolving, pool of partners that are not seen before without any fine-tuning. Population-based training, which approximates such an evolving partner pool, has been proven to provide good zero-shot coordination performance; nevertheless, existing methods are limited by computational resources, mainly focusing on optimizing diversity in small populations while neglecting the potential performance gains from scaling population size. To address this issue, this paper proposes the Scalable Population Training (ScaPT), an efficient RL training framework comprising two key components: a meta-agent that efficiently realizes a population by selectively sharing parameters across agents, and a mutual information regularizer that guarantees population diversity. To empirically validate the effectiveness of ScaPT, this paper evaluates it along with representational frameworks in Hanabi cooperative game and confirms its superiority.