English

Why long model-based rollouts are no reason for bad Q-value estimates

Machine Learning 2024-07-17 v1

Abstract

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.

Keywords

Cite

@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}
}

Comments

Accepted at ESANN 2024

R2 v1 2026-06-28T17:43:06.548Z