Multi-agent assignment via state augmented reinforcement learning
This work addresses multi-agent coordination problems for distributed systems, but it appears incremental as it builds on existing reinforcement learning and gossip communication methods.
The paper tackled the conflicting requirements in multi-agent assignment problems by developing a distributed protocol using state-augmented reinforcement learning, which achieved theoretical feasibility guarantees and was validated in a monitoring experiment.
We address the conflicting requirements of a multi-agent assignment problem through constrained reinforcement learning, emphasizing the inadequacy of standard regularization techniques for this purpose. Instead, we recur to a state augmentation approach in which the oscillation of dual variables is exploited by agents to alternate between tasks. In addition, we coordinate the actions of the multiple agents acting on their local states through these multipliers, which are gossiped through a communication network, eliminating the need to access other agent states. By these means, we propose a distributed multi-agent assignment protocol with theoretical feasibility guarantees that we corroborate in a monitoring numerical experiment.