Unsupervised Reinforcement Learning of Transferable Meta-Skills for Embodied Navigation
This addresses the data efficiency challenge in embodied navigation for robotics and AI applications, though it is incremental as it builds on existing meta-learning and unsupervised methods.
The paper tackles the problem of visual navigation with limited annotated training data by proposing an unsupervised reinforcement learning approach to learn transferable meta-skills, resulting in a 53.34% relative improvement in SPL over baselines in AI2-THOR environments.
Visual navigation is a task of training an embodied agent by intelligently navigating to a target object (e.g., television) using only visual observations. A key challenge for current deep reinforcement learning models lies in the requirements for a large amount of training data. It is exceedingly expensive to construct sufficient 3D synthetic environments annotated with the target object information. In this paper, we focus on visual navigation in the low-resource setting, where we have only a few training environments annotated with object information. We propose a novel unsupervised reinforcement learning approach to learn transferable meta-skills (e.g., bypass obstacles, go straight) from unannotated environments without any supervisory signals. The agent can then fast adapt to visual navigation through learning a high-level master policy to combine these meta-skills, when the visual-navigation-specified reward is provided. Evaluation in the AI2-THOR environments shows that our method significantly outperforms the baseline by 53.34% relatively on SPL, and further qualitative analysis demonstrates that our method learns transferable motor primitives for visual navigation.