Neurosymbolic Transformers for Multi-Agent Communication
This work is significant for multi-agent systems that require efficient communication, especially in bandwidth-limited scenarios, by offering a method to reduce communication overhead without sacrificing performance.
This paper addresses the challenge of inferring communication structures for cooperative multi-agent planning problems while minimizing communication, quantified by the maximum degree of the communication graph. The proposed neurosymbolic transformer algorithm synthesizes policies that generate low-degree communication graphs while maintaining near-optimal performance.
We study the problem of inferring communication structures that can solve cooperative multi-agent planning problems while minimizing the amount of communication. We quantify the amount of communication as the maximum degree of the communication graph; this metric captures settings where agents have limited bandwidth. Minimizing communication is challenging due to the combinatorial nature of both the decision space and the objective; for instance, we cannot solve this problem by training neural networks using gradient descent. We propose a novel algorithm that synthesizes a control policy that combines a programmatic communication policy used to generate the communication graph with a transformer policy network used to choose actions. Our algorithm first trains the transformer policy, which implicitly generates a "soft" communication graph; then, it synthesizes a programmatic communication policy that "hardens" this graph, forming a neurosymbolic transformer. Our experiments demonstrate how our approach can synthesize policies that generate low-degree communication graphs while maintaining near-optimal performance.