Alignment of electron optical beam shaping elements using a convolutional neural network

Machine learning in scanning transmission electron microscopy

“…It is clear that the increasing complexity of these new optical systems will require a more sophisticated control. For this reason, the techniques such as convolutional neural networks 138 as well as more advanced approaches such as Bayesian optimization 139 and reinforcement learning 140 will have an increasing role in the control of the complex optical system and in the alignment of the microscope in general.…”

Machine learning in electron microscopy for advanced nanocharacterization: current developments, available tools and future outlook

“…Importantly, the identification of the features defining the score metric was mostly automated by the reviewed examples. [35][36][37][38][40][41][42][43][44]473 Then, the goal would be to reduce this sparsity or make it less meaningful. One option could be linking each tuneable parameter with a score modification, which would imply a hard encoding of the score.…”

Artificial Neural Network for Automatic Alignment of Electron Optical Devices

“…We will start from the specific case of an orbital angular momentum (OAM) sorter [5], a novel type of miniaturised electron optics that introduces a conformal mapping of the electron beam in order to measure its OAM state spectrum. An ANN, connected to the microscope, is able to reliably improve the alignment of the complex electron optical configuration of the OAM sorter from the observation of a single misaligned OAM spectrum, reaching the same resolution of a skilled user in few seconds (see figure 2, top row).…”