Efficient Model-based Multi-agent Reinforcement Learning via Optimistic Equilibrium Computation

TL;DR

H-MARL is a sample-efficient, model-based multi-agent reinforcement learning algorithm that balances exploration and exploitation by constructing optimistic hallucinated games, achieving fast convergence to equilibrium policies.

Contribution

The paper introduces H-MARL, a novel algorithm that combines confidence intervals and optimistic hallucinated games for efficient multi-agent learning with theoretical guarantees.

Findings

H-MARL learns successful equilibrium policies after few interactions.

It significantly outperforms non-optimistic exploration methods.

The approach is validated on an autonomous driving benchmark.

Abstract

We consider model-based multi-agent reinforcement learning, where the environment transition model is unknown and can only be learned via expensive interactions with the environment. We propose H-MARL (Hallucinated Multi-Agent Reinforcement Learning), a novel sample-efficient algorithm that can efficiently balance exploration, i.e., learning about the environment, and exploitation, i.e., achieve good equilibrium performance in the underlying general-sum Markov game. H-MARL builds high-probability confidence intervals around the unknown transition model and sequentially updates them based on newly observed data. Using these, it constructs an optimistic hallucinated game for the agents for which equilibrium policies are computed at each round. We consider general statistical models (e.g., Gaussian processes, deep ensembles, etc.) and policy classes (e.g., deep neural networks), and theoretically analyze our approach by bounding the agents' dynamic regret. Moreover, we provide a convergence rate to the equilibria of the underlying Markov game. We demonstrate our approach experimentally on an autonomous driving simulation benchmark. H-MARL learns successful equilibrium policies after a few interactions with the environment and can significantly improve the performance compared to non-optimistic exploration methods.