Model-based Reinforcement Learning: A Survey

TL;DR

This survey comprehensively reviews model-based reinforcement learning, covering dynamics model learning challenges, planning integration strategies, and potential benefits, providing a broad conceptual overview of combining planning and learning in MDP optimization.

Contribution

It systematically categorizes approaches to dynamics modeling and planning integration in model-based RL, highlighting challenges and potential benefits, and connects related RL fields.

Findings

Detailed categorization of dynamics model learning approaches

Analysis of planning and learning integration strategies

Discussion of implicit model-based RL and its advantages

Abstract

Sequential decision making, commonly formalized as Markov Decision Process (MDP) optimization, is a important challenge in artificial intelligence. Two key approaches to this problem are reinforcement learning (RL) and planning. This paper presents a survey of the integration of both fields, better known as model-based reinforcement learning. Model-based RL has two main steps. First, we systematically cover approaches to dynamics model learning, including challenges like dealing with stochasticity, uncertainty, partial observability, and temporal abstraction. Second, we present a systematic categorization of planning-learning integration, including aspects like: where to start planning, what budgets to allocate to planning and real data collection, how to plan, and how to integrate planning in the learning and acting loop. After these two sections, we also discuss implicit model-based RL as an end-to-end alternative for model learning and planning, and we cover the potential benefits of model-based RL. Along the way, the survey also draws connections to several related RL fields, like hierarchical RL and transfer learning. Altogether, the survey presents a broad conceptual overview of the combination of planning and learning for MDP optimization.