Organizers: Maxim Ziatdinov, Rama Vasudevan, Debangshu Mukherjee, and Sergei V. Kalinin
The virtual school on AI for Atoms: How to Machine Learn STEM, held December 7-10, 2020, combined invited and contributed presentations on forefront ML applications in Scanning Transmission Electron Microscopy (STEM), Electron Energy Loss Spectroscopy (EELS), and 4D STEM, as well as for physics and chemistry extraction from STEM data sets. It featured tutorials on recent developments in ML analysis of mesoscopic and atomically resolved images and spectroscopy in STEM, including classical graph analysis of STEM data, deep convolutional neural networks for feature identifications, symmetry-invariant autoencoders, and Gaussian Processing based super-resolution imaging and image reconstruction.