Learning to Navigate Using Mid-Level Visual Priors
This addresses the challenge of improving sample efficiency and generalization in reinforcement learning for robotics and autonomous systems, though it is incremental by building on existing representation learning methods.
The paper tackles the problem of how visual priors assist in learning motor tasks like navigation, showing that integrating mid-level vision skills (e.g., distance estimators) into reinforcement learning leads to policies that learn faster, generalize better, and achieve higher final performance compared to baseline methods.
How much does having visual priors about the world (e.g. the fact that the world is 3D) assist in learning to perform downstream motor tasks (e.g. navigating a complex environment)? What are the consequences of not utilizing such visual priors in learning? We study these questions by integrating a generic perceptual skill set (a distance estimator, an edge detector, etc.) within a reinforcement learning framework (see Fig. 1). This skill set ("mid-level vision") provides the policy with a more processed state of the world compared to raw images. Our large-scale study demonstrates that using mid-level vision results in policies that learn faster, generalize better, and achieve higher final performance, when compared to learning from scratch and/or using state-of-the-art visual and non-visual representation learning methods. We show that conventional computer vision objectives are particularly effective in this regard and can be conveniently integrated into reinforcement learning frameworks. Finally, we found that no single visual representation was universally useful for all downstream tasks, hence we computationally derive a task-agnostic set of representations optimized to support arbitrary downstream tasks.