Ma Desheng, Zeltmann Steven E, Zhang Chenyu, Baraissov Zhaslan, Shao Yu-Tsun, Duncan Cameron, Maxson Jared, Edelen Auralee, Muller David A
School of Applied and Engineering Physics, Cornell University, Ithaca, NY 14853, USA.
Platform for the Accelerated Realization, Analysis, and Discovery of Interface Materials and School of Applied and Engineering Physics, Cornell University, Ithaca, NY 14853, USA.
Ultramicroscopy. 2025 Jul;273:114138. doi: 10.1016/j.ultramic.2025.114138. Epub 2025 Apr 4.
Aberration-corrected Scanning Transmission Electron Microscopy (STEM) has become an essential tool in understanding materials at the atomic scale. However, tuning the aberration corrector to produce a sub-Ångström probe is a complex and time-costly procedure, largely due to the difficulty of precisely measuring the optical state of the system. When measurements are both costly and noisy, Bayesian methods provide rapid and efficient optimization. To this end, we develop a Bayesian approach to fully automate the process by minimizing a new quality metric, beam emittance, which is shown to be equivalent to performing aberration correction. In part I, we derived several important properties of the beam emittance metric and trained a deep neural network to predict beam emittance growth from a single Ronchigram. Here we use this as the black box function for Bayesian optimization and demonstrate automated tuning of simulated and real electron microscopes. We explore different surrogate functions for the Bayesian optimizer and implement a deep neural network kernel to effectively learn the interactions between different control channels without the need to explicitly measure a full set of aberration coefficients. Both simulation and experimental results show the proposed method outperforms conventional approaches by achieving a better optical state with a higher convergence rate.
像差校正扫描透射电子显微镜（STEM）已成为在原子尺度上理解材料的重要工具。然而，调整像差校正器以产生亚埃探针是一个复杂且耗时的过程，这主要是由于精确测量系统光学状态存在困难。当测量既昂贵又有噪声时，贝叶斯方法能提供快速有效的优化。为此，我们开发了一种贝叶斯方法，通过最小化一个新的质量指标——束发射度，来使该过程完全自动化，结果表明这等同于进行像差校正。在第一部分中，我们推导了束发射度指标的几个重要特性，并训练了一个深度神经网络，以从单个 Ronchigram 预测束发射度增长。在此，我们将此用作贝叶斯优化的黑箱函数，并展示了对模拟和实际电子显微镜的自动调谐。我们探索了贝叶斯优化器的不同替代函数，并实现了一个深度神经网络内核，以有效地学习不同控制通道之间的相互作用，而无需明确测量全套像差系数。模拟和实验结果均表明，所提出的方法通过以更高的收敛速率实现更好的光学状态，优于传统方法。
