CVLGROMar 11, 2022

Masked Visual Pre-training for Motor Control

arXiv:2203.06173v1308 citationsh-index: 156
Originality Highly original
AI Analysis

This addresses the problem of efficient and scalable motor control learning for robotics without relying on labels or expert demonstrations, representing a novel approach in the field.

The paper tackles learning motor control tasks from pixels by using self-supervised visual pre-training from real-world images, achieving up to 80% higher success rates than supervised encoders and sometimes matching oracle state performance.

This paper shows that self-supervised visual pre-training from real-world images is effective for learning motor control tasks from pixels. We first train the visual representations by masked modeling of natural images. We then freeze the visual encoder and train neural network controllers on top with reinforcement learning. We do not perform any task-specific fine-tuning of the encoder; the same visual representations are used for all motor control tasks. To the best of our knowledge, this is the first self-supervised model to exploit real-world images at scale for motor control. To accelerate progress in learning from pixels, we contribute a benchmark suite of hand-designed tasks varying in movements, scenes, and robots. Without relying on labels, state-estimation, or expert demonstrations, we consistently outperform supervised encoders by up to 80% absolute success rate, sometimes even matching the oracle state performance. We also find that in-the-wild images, e.g., from YouTube or Egocentric videos, lead to better visual representations for various manipulation tasks than ImageNet images.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes