A Unified View on Solving Objective Mismatch in Model-Based Reinforcement Learning
This work addresses a fundamental challenge in making MBRL agents more effective, but it is incremental as it synthesizes existing research rather than introducing new methods.
The paper tackles the objective mismatch problem in model-based reinforcement learning, where accurate dynamics models do not guarantee good policy performance, by providing a survey and taxonomy of existing solution categories to guide future research.
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 has shown that model predictive accuracy is often not correlated with action quality, tracing the root cause to the objective mismatch between accurate dynamics model learning and policy optimization of rewards. A number of interrelated solution categories to the objective mismatch problem have emerged as MBRL continues to mature as a research area. In this work, we provide an in-depth survey of these solution categories and propose a taxonomy to foster future research.