ROMA: Multi-Agent Reinforcement Learning with Emergent Roles

TL;DR

ROMA introduces a role-oriented MARL framework where roles emerge dynamically, enabling agents to specialize and improve performance on complex tasks like StarCraft II micromanagement.

Contribution

The paper proposes a novel MARL framework with emergent roles using a stochastic role embedding space and regularizers, enhancing adaptability and specialization.

Findings

Learned roles are specialized, dynamic, and identifiable.

ROMA outperforms existing methods on StarCraft II benchmark.

Roles facilitate better coordination and task performance.

Abstract

The role concept provides a useful tool to design and understand complex multi-agent systems, which allows agents with a similar role to share similar behaviors. However, existing role-based methods use prior domain knowledge and predefine role structures and behaviors. In contrast, multi-agent reinforcement learning (MARL) provides flexibility and adaptability, but less efficiency in complex tasks. In this paper, we synergize these two paradigms and propose a role-oriented MARL framework (ROMA). In this framework, roles are emergent, and agents with similar roles tend to share their learning and to be specialized on certain sub-tasks. To this end, we construct a stochastic role embedding space by introducing two novel regularizers and conditioning individual policies on roles. Experiments show that our method can learn specialized, dynamic, and identifiable roles, which help our method push forward the state of the art on the StarCraft II micromanagement benchmark. Demonstrative videos are available at https://sites.google.com/view/romarl/.