Collection 

Machine learning for automated experimentation in scanning transmission electron microscopy (STEM)

Submission status
Closed
Submission deadline

The special Collection “Machine learning for automated experimentation in STEM” explores all aspects of the integration of machine learning (ML) into STEM to transform experimental workflows. This includes real-time data analysis, setting up of data analysis workflows, and facilitating autonomous decision-making in microscopy by leveraging the latest in machine learning methodologies. The combination of advances in the hardware of STEM along with the software and algorithms afforded by modern day ML methods can lead to significant advances in materials science and nanotechnology. This Collection welcomes the following topics, including, but not limited to:   

1. ML-based analysis methods for imaging and spectroscopic data in STEM: for both real-time and offline data analysis.
2. Automated STEM workflows: data pipelines, software and hardware ecosystems and enabling instrumentation to deliver automated and autonomous STEM platforms.
3. Simulations for STEM: STEM simulations of imaging and spectroscopy that can be compared with experimental observations or can be used to train models that can be used for data analysis tasks.
4. Decision making for autonomous STEM: Algorithms and methods, such as Bayesian optimization and reinforcement learning, to take decisions on the fly in STEM in real-time for real-time control over STEM instrumentation for achieving materials manipulation, instrument control, etc.
5. Theory-experiment coupling: Papers that explore theory-experiment coupling, especially those where these are done in real-time, for example where theoretical models are constructed and fit in real time to drive navigation in the experimental domain, and vice versa.

machine learning

Editors

Please follow the steps detailed on this page, to prepare your manuscript for submission. Submissions are handled via our online submission system. When filling out the manuscript information, in the "Subject Terms" tab, select this Collection from the alphabetical list. Authors should express their interest in the Collection in their cover letter.