MoVie: Visual Model-Based Policy Adaptation for View Generalization
It addresses a key challenge for robotics applications where agents must generalize from limited training views to real-world scenarios, representing a strong specific gain in a domain-specific area.
The paper tackles the problem of view generalization in visual reinforcement learning by proposing MoVie, a method that adapts model-based policies to unseen views without reward signals or training modifications, achieving relative improvements of 33%, 86%, and 152% across 18 tasks in DMControl, xArm, and Adroit.
Visual Reinforcement Learning (RL) agents trained on limited views face significant challenges in generalizing their learned abilities to unseen views. This inherent difficulty is known as the problem of $\textit{view generalization}$. In this work, we systematically categorize this fundamental problem into four distinct and highly challenging scenarios that closely resemble real-world situations. Subsequently, we propose a straightforward yet effective approach to enable successful adaptation of visual $\textbf{Mo}$del-based policies for $\textbf{Vie}$w generalization ($\textbf{MoVie}$) during test time, without any need for explicit reward signals and any modification during training time. Our method demonstrates substantial advancements across all four scenarios encompassing a total of $\textbf{18}$ tasks sourced from DMControl, xArm, and Adroit, with a relative improvement of $\mathbf{33}$%, $\mathbf{86}$%, and $\mathbf{152}$% respectively. The superior results highlight the immense potential of our approach for real-world robotics applications. Videos are available at https://yangsizhe.github.io/MoVie/ .