ClusterComm: Discrete Communication in Decentralized MARL using Internal Representation Clustering
This addresses communication challenges in MARL for collaborative tasks, but it is incremental as it builds on existing methods with a novel discretization approach.
The paper tackles the problem of improving communication in decentralized multi-agent reinforcement learning by introducing ClusterComm, which uses clustering on internal representations to generate discrete messages, resulting in performance that outperforms no communication and competes with continuous communication.
In the realm of Multi-Agent Reinforcement Learning (MARL), prevailing approaches exhibit shortcomings in aligning with human learning, robustness, and scalability. Addressing this, we introduce ClusterComm, a fully decentralized MARL framework where agents communicate discretely without a central control unit. ClusterComm utilizes Mini-Batch-K-Means clustering on the last hidden layer's activations of an agent's policy network, translating them into discrete messages. This approach outperforms no communication and competes favorably with unbounded, continuous communication and hence poses a simple yet effective strategy for enhancing collaborative task-solving in MARL.