Communicating Unexpectedness for Out-of-Distribution Multi-Agent Reinforcement Learning

TL;DR

This paper introduces Unexpected Encoding Scheme, a decentralized multi-agent reinforcement learning method where agents communicate surprising environmental aspects to improve adaptation to unforeseen situations, enhancing robustness in dynamic environments.

Contribution

The paper presents a novel communication approach for multi-agent RL that encodes unexpected environmental changes, enabling better adaptation to out-of-distribution scenarios.

Findings

Supports robust adaptation to dynamic environments

Improves out-of-distribution generalization in multi-agent settings

Demonstrates effectiveness in multi-robot warehouse tasks

Abstract

Applying multi-agent reinforcement learning methods to realistic settings is challenging as it may require the agents to quickly adapt to unexpected situations that are rarely or never encountered in training. Recent methods for generalization to such out-of-distribution settings are limited to more specific, restricted instances of distribution shifts. To tackle adaptation to distribution shifts, we propose Unexpected Encoding Scheme, a novel decentralized multi-agent reinforcement learning algorithm where agents communicate "unexpectedness," the aspects of the environment that are surprising. In addition to a message yielded by the original reward-driven communication, each agent predicts the next observation based on previous experience, measures the discrepancy between the prediction and the actually encountered observation, and encodes this discrepancy as a message. Experiments on multi-robot warehouse environment support that our proposed method adapts robustly to dynamically changing training environments as well as out-of-distribution environment.