MAMBA: an Effective World Model Approach for Meta-Reinforcement Learning

TL;DR

This paper introduces MAMBA, a model-based meta-reinforcement learning approach that significantly improves sample efficiency and performance on benchmark domains, advancing towards real-world applicability.

Contribution

MAMBA combines state-of-the-art model-based and meta-RL techniques to enhance sample efficiency and scalability in meta-RL tasks, especially in high-dimensional environments.

Findings

Achieves up to 15x better sample efficiency.

Attains higher returns on benchmark domains.

Performs well with minimal hyperparameter tuning.

Abstract

Meta-reinforcement learning (meta-RL) is a promising framework for tackling challenging domains requiring efficient exploration. Existing meta-RL algorithms are characterized by low sample efficiency, and mostly focus on low-dimensional task distributions. In parallel, model-based RL methods have been successful in solving partially observable MDPs, of which meta-RL is a special case. In this work, we leverage this success and propose a new model-based approach to meta-RL, based on elements from existing state-of-the-art model-based and meta-RL methods. We demonstrate the effectiveness of our approach on common meta-RL benchmark domains, attaining greater return with better sample efficiency (up to $15\times$) while requiring very little hyperparameter tuning. In addition, we validate our approach on a slate of more challenging, higher-dimensional domains, taking a step towards real-world generalizing agents.