Interferobot: Aligning an Optical Interferometer by a Reinforcement Learning Agent
Abstract
Limitations in acquiring training data restrict potential applications of deep reinforcement learning (RL) methods to the training of real-world robots. Here we train an RL agent to align a Mach-Zehnder interferometer, which is an essential part of many optical experiments, based on images of interference fringes acquired by a monocular camera. The agent is trained in a simulated environment, without any hand-coded features or a priori information about the physics, and subsequently transferred to a physical interferometer. Thanks to a set of domain randomizations simulating uncertainties in physical measurements, the agent successfully aligns this interferometer without any fine-tuning, achieving a performance level of a human expert.
Cite
Text
Sorokin et al. "Interferobot: Aligning an Optical Interferometer by a Reinforcement Learning Agent." Neural Information Processing Systems, 2020.Markdown
[Sorokin et al. "Interferobot: Aligning an Optical Interferometer by a Reinforcement Learning Agent." Neural Information Processing Systems, 2020.](https://mlanthology.org/neurips/2020/sorokin2020neurips-interferobot/)BibTeX
@inproceedings{sorokin2020neurips-interferobot,
title = {{Interferobot: Aligning an Optical Interferometer by a Reinforcement Learning Agent}},
author = {Sorokin, Dmitry and Ulanov, Alexander and Sazhina, Ekaterina and Lvovsky, Alexander},
booktitle = {Neural Information Processing Systems},
year = {2020},
url = {https://mlanthology.org/neurips/2020/sorokin2020neurips-interferobot/}
}