Contrasting Centralized and Decentralized Critics in Multi-Agent Reinforcement Learning
This work clarifies the trade-offs between centralized and decentralized critics for multi-agent reinforcement learning algorithm designers, addressing a common misconception in the field.
This paper analyzes the implications of using centralized versus decentralized critics in multi-agent reinforcement learning, a common design choice in actor-critic methods. It reveals that centralized critics are not universally superior, but rather both approaches have distinct advantages and disadvantages.
Centralized Training for Decentralized Execution, where agents are trained offline using centralized information but execute in a decentralized manner online, has gained popularity in the multi-agent reinforcement learning community. In particular, actor-critic methods with a centralized critic and decentralized actors are a common instance of this idea. However, the implications of using a centralized critic in this context are not fully discussed and understood even though it is the standard choice of many algorithms. We therefore formally analyze centralized and decentralized critic approaches, providing a deeper understanding of the implications of critic choice. Because our theory makes unrealistic assumptions, we also empirically compare the centralized and decentralized critic methods over a wide set of environments to validate our theories and to provide practical advice. We show that there exist misconceptions regarding centralized critics in the current literature and show that the centralized critic design is not strictly beneficial, but rather both centralized and decentralized critics have different pros and cons that should be taken into account by algorithm designers.