English

Pre-training of Deep RL Agents for Improved Learning under Domain Randomization

Machine Learning 2021-04-30 v1 Artificial Intelligence Robotics

Abstract

Visual domain randomization in simulated environments is a widely used method to transfer policies trained in simulation to real robots. However, domain randomization and augmentation hamper the training of a policy. As reinforcement learning struggles with a noisy training signal, this additional nuisance can drastically impede training. For difficult tasks it can even result in complete failure to learn. To overcome this problem we propose to pre-train a perception encoder that already provides an embedding invariant to the randomization. We demonstrate that this yields consistently improved results on a randomized version of DeepMind control suite tasks and a stacking environment on arbitrary backgrounds with zero-shot transfer to a physical robot.

Keywords

Cite

@article{arxiv.2104.14386,
  title  = {Pre-training of Deep RL Agents for Improved Learning under Domain Randomization},
  author = {Artemij Amiranashvili and Max Argus and Lukas Hermann and Wolfram Burgard and Thomas Brox},
  journal= {arXiv preprint arXiv:2104.14386},
  year   = {2021}
}
R2 v1 2026-06-24T01:38:09.115Z