Guardado en:
Detalles Bibliográficos
Autores principales: Brown, Davis, Nizinski, Cody, Shapiro, Madelyn, Fallon, Corey, Yin, Tianzhixi, Kvinge, Henry, Tu, Jonathan H.
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2407.15756
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911964191522816
author Brown, Davis
Nizinski, Cody
Shapiro, Madelyn
Fallon, Corey
Yin, Tianzhixi
Kvinge, Henry
Tu, Jonathan H.
author_facet Brown, Davis
Nizinski, Cody
Shapiro, Madelyn
Fallon, Corey
Yin, Tianzhixi
Kvinge, Henry
Tu, Jonathan H.
contents Deep learning still struggles with certain kinds of scientific data. Notably, pretraining data may not provide coverage of relevant distribution shifts (e.g., shifts induced via the use of different measurement instruments). We consider deep learning models trained to classify the synthesis conditions of uranium ore concentrates (UOCs) and show that model editing is particularly effective for improving generalization to distribution shifts common in this domain. In particular, model editing outperforms finetuning on two curated datasets comprising of micrographs taken of U$_{3}$O$_{8}$ aged in humidity chambers and micrographs acquired with different scanning electron microscopes, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2407_15756
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Model editing for distribution shifts in uranium oxide morphological analysis
Brown, Davis
Nizinski, Cody
Shapiro, Madelyn
Fallon, Corey
Yin, Tianzhixi
Kvinge, Henry
Tu, Jonathan H.
Machine Learning
Artificial Intelligence
Deep learning still struggles with certain kinds of scientific data. Notably, pretraining data may not provide coverage of relevant distribution shifts (e.g., shifts induced via the use of different measurement instruments). We consider deep learning models trained to classify the synthesis conditions of uranium ore concentrates (UOCs) and show that model editing is particularly effective for improving generalization to distribution shifts common in this domain. In particular, model editing outperforms finetuning on two curated datasets comprising of micrographs taken of U$_{3}$O$_{8}$ aged in humidity chambers and micrographs acquired with different scanning electron microscopes, respectively.
title Model editing for distribution shifts in uranium oxide morphological analysis
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2407.15756