Towards Augmented Microscopy with Reinforcement Learning-Enhanced Workflows

TL;DR

This paper demonstrates how reinforcement learning can automate electron microscope operations by developing a virtual environment for training and successfully deploying a model on actual equipment, advancing microscopy workflows.

Contribution

It introduces a virtual RL environment for microscopy automation and validates its effectiveness through real-world deployment, showing minimal training needed for accurate alignment.

Findings

Robust alignment accuracy across hyperparameters

Successful deployment on actual microscope

Minimal training required for convergence

Abstract

Here, we report a case study implementation of reinforcement learning (RL) to automate operations in the scanning transmission electron microscopy (STEM) workflow. To do so, we design a virtual, prototypical RL environment to test and develop a network to autonomously align the electron beam without prior knowledge. Using this simulator, we evaluate the impact of environment design and algorithm hyperparameters on alignment accuracy and learning convergence, showing robust convergence across a wide hyperparameter space. Additionally, we deploy a successful model on the microscope to validate the approach and demonstrate the value of designing appropriate virtual environments. Consistent with simulated results, the on-microscope RL model achieves convergence to the goal alignment after minimal training. Overall, the results highlight that by taking advantage of RL, microscope operations can be automated without the need for extensive algorithm design, taking another step towards augmenting electron microscopy with machine learning methods.