This paper explores the use of model-based offline reinforcement learning with long model rollouts. While some literature criticizes this approach due to compounding errors, many practitioners have found success in real-world applications. The paper aims to demonstrate that long rollouts do not necessarily result in exponentially growing errors and can actually produce better Q-value estimates than model-free methods. These findings can potentially enhance reinforcement learning techniques.
@article{arxiv.2407.11751,
title = {Why long model-based rollouts are no reason for bad Q-value estimates},
author = {Philipp Wissmann and Daniel Hein and Steffen Udluft and Volker Tresp},
journal= {arXiv preprint arXiv:2407.11751},
year = {2024}
}