Mic-hackathon 2024：基于机器学习的电子显微镜与扫描探针显微镜黑客马拉松
材料科学
2025-06-30 v2 机器学习
仪器与探测器
摘要
显微镜是获取材料在纳米和原子尺度上结构与功能信息的主要来源。生成的数据常为结构化、包含丰富的元数据和样本历史，虽然在细节或格式上不始终一致。 major 资助机构推广的数据管理计划（DMP）促进了数据的保存与访问。然而，由于缺乏标准化的代码生态系统、基准测试和集成策略，仍难从数据中提取见解。结果是数据使用效率低下，分析耗时。除了数据获取后分析外，来自 major 显微镜制造商的新型 API 使实时基于机器学习的分析成为可能，支持自动决策和由 ML 代理控制的显微镜操作。然而，机器学习与显微镜社区之间仍存在鸿沟，限制了这些方法对物理学、材料发现和优化的影响。黑客马拉松有助于搭建这一桥梁，通过鼓励机器学习研究人员与显微镜专家合作，推动将机器学习应用于显微镜的创新解决方案的开发，并为仪器、材料科学和应用机器学习的未来 workforce 做准备。本次黑客马拉松产生了基准数据集和显微镜的数字孪生体，以支持社区增长和标准化工作流程。所有相关代码均 available at GitHub: https://github.com/KalininGroup/Mic-hackathon-2024-codes-publication/tree/1.0.0.1
引用
@article{arxiv.2506.08423,
title = {Mic-hackathon 2024: Hackathon on Machine Learning for Electron and Scanning Probe Microscopy},
author = {Utkarsh Pratiush and Austin Houston and Kamyar Barakati and Aditya Raghavan and Dasol Yoon and Harikrishnan KP and Zhaslan Baraissov and Desheng Ma and Samuel S. Welborn and Mikolaj Jakowski and Shawn-Patrick Barhorst and Alexander J. Pattison and Panayotis Manganaris and Sita Sirisha Madugula and Sai Venkata Gayathri Ayyagari and Vishal Kennedy and Ralph Bulanadi and Michelle Wang and Kieran J. Pang and Ian Addison-Smith and Willy Menacho and Horacio V. Guzman and Alexander Kiefer and Nicholas Furth and Nikola L. Kolev and Mikhail Petrov and Viktoriia Liu and Sergey Ilyev and Srikar Rairao and Tommaso Rodani and Ivan Pinto-Huguet and Xuli Chen and Josep Cruañes and Marta Torrens and Jovan Pomar and Fanzhi Su and Pawan Vedanti and Zhiheng Lyu and Xingzhi Wang and Lehan Yao and Amir Taqieddin and Forrest Laskowski and Xiangyu Yin and Yu-Tsun Shao and Benjamin Fein-Ashley and Yi Jiang and Vineet Kumar and Himanshu Mishra and Yogesh Paul and Adib Bazgir and Rama chandra Praneeth Madugula and Yuwen Zhang and Pravan Omprakash and Jian Huang and Eric Montufar-Morales and Vivek Chawla and Harshit Sethi and Jie Huang and Lauri Kurki and Grace Guinan and Addison Salvador and Arman Ter-Petrosyan and Madeline Van Winkle and Steven R. Spurgeon and Ganesh Narasimha and Zijie Wu and Richard Liu and Yongtao Liu and Boris Slautin and Andrew R Lupini and Rama Vasudevan and Gerd Duscher and Sergei V. Kalinin},
journal= {arXiv preprint arXiv:2506.08423},
year = {2025}
}