A Unified View on Solving Objective Mismatch in Model-Based Reinforcement Learning

arXiv:2310.06253v2 Announce Type: replace
Abstract: Model-based Reinforcement Learning (MBRL) aims to make agents more sample-efficient, adaptive, and explainable by learning an explicit model of the environment. While the capabilities of MBRL agents have significantly improved in recent years, how to best learn the model is still an unresolved question. The majority of MBRL algorithms aim at training the model to make accurate predictions about the environment and subsequently using the model to determine the most rewarding actions. However, recent research …

abstract agents arxiv capabilities cs.lg environment learn question reinforcement reinforcement learning sample the environment type view