[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-290248-105":59,"doc-detail-290248-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","simulation-and-machine-learning-for-hrtem-analysis-thesisdissertation","Simulation and Machine Learning for HRTEM Analysis - Thesis/Dissertation","","Transmission electron microscopy is a powerful tool for studying nanomaterials at atomic scale, and recent machine learning advances promise quantitative, information-rich characterization. Limited suitable experimental datasets make electron-microscopy model development difficult. This dissertation trains machine learning models for HRTEM analysis purely via simulation, using random structure generation and multislice simulation. It focuses on semantic segmentation of HRTEM nanoparticle images on amorphous substrates, evaluates high-throughput neural network training, studies domain adaptation via transfer learning, and introduces an intrinsic-information framework to predict out-of-distribution generalization.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/simulation-and-machine-learning-for-hrtem-analysis-thesisdissertation/290248/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/simulation-and-machine-learning-for-hrtem-analysis-thesisdissertation/290248.png","ImageObject",300,407,{"name":92,"@type":93},"Finn","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-17","2026-09-17",true,{"@type":102,"interactionType":103,"userInteractionCount":44},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"Why does the thesis focus on simulation-based training for HRTEM models?","Question",{"text":112,"@type":113},"Experimental microscopy data suitable for training is limited, making standard machine learning development difficult or infeasible. The thesis trains models entirely via simulation to overcome this constraint.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What is the primary machine learning task addressed in the dissertation?",{"text":117,"@type":113},"The work focuses on semantic segmentation of high-resolution transmission electron microscopy (HRTEM) images of nanoparticles on amorphous substrates.",{"name":119,"@type":110,"acceptedAnswer":120},"How does the thesis study robustness and generalization beyond the training domain?",{"text":121,"@type":113},"It examines domain adaptation using transfer learning and proposes a framework to predict out-of-distribution generalization based on intrinsic information transfer characteristics of the imaging datasets.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},290248,1789641128,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":44,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":62,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":129,"read_time":143,"preview_page_count":4},34359740700684,"https://ap-avatar.wpscdn.com/avatar/1f400023980c374ae676?_k=1777273430885731487","UC Berkeley  \nUC Berkeley Electronic Theses and Dissertations  \nTitle  \nSimulation and Machine Learning for HRTEM Analysis  \nPermalink  \n[https://escholarship.org/uc/item/8bg616k6](https://escholarship.org/uc/item/8bg616k6)  \nISBN  \n9798276023403  \nAuthor  \nRangel DaCosta, Luis  \nPublication Date  \n2025-12-06  \nPeer reviewed|Thesis/dissertation  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nSimulation and Machine Learning for HRTEM Analysis  \nBy  \nLuis Rangel DaCosta  \nA dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy  \nin  \nEngineering–Materials Science and Engineering  \nin the  \nGraduate Division  \nof the  \nUniversity of California, Berkeley  \nCommittee in charge:  \nProfessor Mary Scott, Chair Professor Andy Minor Professor Joshua Bloom  \nFall 2025  \nSimulation and Machine Learning for HRTEM Analysis  \nCopyright 2025  \nby  \nLuis Rangel DaCosta  \n1  \nAbstract  \nSimulation and Machine Learning for HRTEM Analysis  \nby  \nLuis Rangel DaCosta  \nDoctor of Philosophy in Engineering–Materials Science and Engineering  \nUniversity of California, Berkeley  \nProfessor Mary Scott, Chair  \nTransmission electron microscopy is one of the most powerful characterization techniques available for studying nanomaterials systems at the atomic scale. Recent machine learning advances offer to revolutionize our characterization workflows with unparalleled quantitative analysis and information extraction abilities. Such advanced capabilities could vastly increase the scope and throughput of electron microscopy experiments, but due to the limited quantity of experimental microscopy data which is suitable for training machine learning models, developing new machine learning capabilities for electron microscopy applications can be difficult or practically infeasible. This thesis develops methodologies for training machine learning models for electron microscopy applications entirely via simulation using random structure generation and multislice simulation techniques, focusing on semantic segmentation of high-resolution transmission electron microscopy (HRTEM) images of nanoparticles on amorphous substrates. Across a series of high-throughput neural network training studies, we demonstrate the efficacy of simulation training for HRTEM analysis with machine learning models, study the domain adaptation of neural network models via transfer learning, and develop a framework to predict the out-of-distribution generalization behavior of HRTEM analysis models via the intrinsic information transfer characteristics of imaging datasets.  \ni  \nI would guess most doctorates are not truly earned on their own; mine certainly was not. I am so deeply grateful for the support of many, for those in my life who without question believed in what I thought I could do and what I had to say, unwaveringly, for years! while I was squirreled away, mostly writing code and often saying very little, hoping to make new things come true. Without community, I am not so sure all of this would have been worth the hassle, and certainly would not have been nearly as fulfilling and enjoyable. It has been so important for me to find myself in places that truly nourish; that are energizing and thriving; that help you feel that all the things you wish to set out for are in reach. In many ways, Berkeley has been one of those places for me—in my experience, these places are always such because of the people who are there—and I feel that any doctorate I have earned has been earned only with community.  \nThere are so many to whom I would like to explicitly extend my gratitude! First and foremost, to my friends and colleagues in the Scott Lab—thank you for keeping things interesting, exciting, fun, and special (and for tolerating all of my questions about experiments I myself never did); and to Mary herself, for all of her guidance, encouragement, and mentorship, and for giving ","cbCaio8vrfUvVr3J","https://ap.wps.com/l/cbCaio8vrfUvVr3J","pdf",28752304,"English","# Introduction\n# Multislice simulation of HRTEM images\n## The multislice algorithm in a nut shell\n## Calculation of atomic scattering potentials\n## Objective lens aberrations\n## Coherence of the electron beam\n## Inelastic scattering effects\n## Camera artifacts and noise effects\n# Generation of realistic atomic structures","[{\"question\":\"Why does the thesis focus on simulation-based training for HRTEM models?\",\"answer\":\"Experimental microscopy data suitable for training is limited, making standard machine learning development difficult or infeasible. The thesis trains models entirely via simulation to overcome this constraint.\"},{\"question\":\"What is the primary machine learning task addressed in the dissertation?\",\"answer\":\"The work focuses on semantic segmentation of high-resolution transmission electron microscopy (HRTEM) images of nanoparticles on amorphous substrates.\"},{\"question\":\"How does the thesis study robustness and generalization beyond the training domain?\",\"answer\":\"It examines domain adaptation using transfer learning and proposes a framework to predict out-of-distribution generalization based on intrinsic information transfer characteristics of the imaging datasets.\"}]","Simulation and Machine Learning for HRTEM Analysis - Thesis/Dissertation | PDF",265]