Efficient Model-based Multi-agent Reinforcement Learning via Optimistic Equilibrium Computation
This addresses the challenge of expensive environment interactions in multi-agent systems, offering a novel approach with theoretical guarantees and experimental gains, though it appears incremental as it builds on existing model-based and optimistic methods.
The paper tackles the problem of sample-efficient model-based multi-agent reinforcement learning in unknown environments by proposing H-MARL, which balances exploration and exploitation using optimistic equilibrium computation, achieving successful equilibrium policies after few interactions and significantly outperforming non-optimistic methods in an autonomous driving simulation.
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.