Modeling Sensorimotor Coordination as Multi-Agent Reinforcement Learning with Differentiable Communication
This work addresses the limitation of homogeneous agents in multi-agent systems, potentially benefiting fields like neuroscience and AI for understanding brain function and building intelligent systems, though it appears incremental as it builds on existing differentiable communication methods.
The paper tackled the problem of applying multi-agent reinforcement learning to complex tasks by proposing a modular framework of cooperating heterogeneous agents, and demonstrated proof-of-concept for modeling sensorimotor coordination in the neocortex.
Multi-agent reinforcement learning has shown promise on a variety of cooperative tasks as a consequence of recent developments in differentiable inter-agent communication. However, most architectures are limited to pools of homogeneous agents, limiting their applicability. Here we propose a modular framework for learning complex tasks in which a traditional monolithic agent is framed as a collection of cooperating heterogeneous agents. We apply this approach to model sensorimotor coordination in the neocortex as a multi-agent reinforcement learning problem. Our results demonstrate proof-of-concept of the proposed architecture and open new avenues for learning complex tasks and for understanding functional localization in the brain and future intelligent systems.