Learning to Schedule Communication in Multi-agent Reinforcement Learning

TL;DR

This paper introduces SchedNet, a multi-agent reinforcement learning framework that learns to schedule communication among agents with limited bandwidth, improving coordination in tasks like navigation and predator-prey.

Contribution

The paper presents a novel deep reinforcement learning approach for communication scheduling in multi-agent systems with bandwidth constraints, enabling adaptive and importance-based message broadcasting.

Findings

SchedNet outperforms baselines by 32-43% in cooperative tasks.

Learned scheduling improves coordination under bandwidth limitations.

Agents effectively prioritize important information for communication.

Abstract

Many real-world reinforcement learning tasks require multiple agents to make sequential decisions under the agents' interaction, where well-coordinated actions among the agents are crucial to achieve the target goal better at these tasks. One way to accelerate the coordination effect is to enable multiple agents to communicate with each other in a distributed manner and behave as a group. In this paper, we study a practical scenario when (i) the communication bandwidth is limited and (ii) the agents share the communication medium so that only a restricted number of agents are able to simultaneously use the medium, as in the state-of-the-art wireless networking standards. This calls for a certain form of communication scheduling. In that regard, we propose a multi-agent deep reinforcement learning framework, called SchedNet, in which agents learn how to schedule themselves, how to encode the messages, and how to select actions based on received messages. SchedNet is capable of deciding which agents should be entitled to broadcasting their (encoded) messages, by learning the importance of each agent's partially observed information. We evaluate SchedNet against multiple baselines under two different applications, namely, cooperative communication and navigation, and predator-prey. Our experiments show a non-negligible performance gap between SchedNet and other mechanisms such as the ones without communication and with vanilla scheduling methods, e.g., round robin, ranging from 32% to 43%.