Off-the-shelf deep learning is not enough, and requires parsimony, Bayesianity, and causality

Abstract Deep neural networks (‘deep learning’) have emerged as a technology of choice to tackle problems in speech recognition, computer vision, finance, etc. However, adoption of deep learning in physical domains brings substantial challenges stemming from the correlative nature of deep learning m...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Rama K. Vasudevan, Maxim Ziatdinov, Lukas Vlcek, Sergei V. Kalinin
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
Acceso en línea:https://doaj.org/article/71fb8b9fb2e94aae97f4e00701a335d9
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:71fb8b9fb2e94aae97f4e00701a335d9
record_format dspace
spelling oai:doaj.org-article:71fb8b9fb2e94aae97f4e00701a335d92021-12-02T13:57:35ZOff-the-shelf deep learning is not enough, and requires parsimony, Bayesianity, and causality10.1038/s41524-020-00487-02057-3960https://doaj.org/article/71fb8b9fb2e94aae97f4e00701a335d92021-01-01T00:00:00Zhttps://doi.org/10.1038/s41524-020-00487-0https://doaj.org/toc/2057-3960Abstract Deep neural networks (‘deep learning’) have emerged as a technology of choice to tackle problems in speech recognition, computer vision, finance, etc. However, adoption of deep learning in physical domains brings substantial challenges stemming from the correlative nature of deep learning methods compared to the causal, hypothesis driven nature of modern science. We argue that the broad adoption of Bayesian methods incorporating prior knowledge, development of solutions with incorporated physical constraints and parsimonious structural descriptors and generative models, and ultimately adoption of causal models, offers a path forward for fundamental and applied research.Rama K. VasudevanMaxim ZiatdinovLukas VlcekSergei V. KalininNature PortfolioarticleMaterials of engineering and construction. Mechanics of materialsTA401-492Computer softwareQA76.75-76.765ENnpj Computational Materials, Vol 7, Iss 1, Pp 1-6 (2021)
institution DOAJ
collection DOAJ
language EN
topic Materials of engineering and construction. Mechanics of materials
TA401-492
Computer software
QA76.75-76.765
spellingShingle Materials of engineering and construction. Mechanics of materials
TA401-492
Computer software
QA76.75-76.765
Rama K. Vasudevan
Maxim Ziatdinov
Lukas Vlcek
Sergei V. Kalinin
Off-the-shelf deep learning is not enough, and requires parsimony, Bayesianity, and causality
description Abstract Deep neural networks (‘deep learning’) have emerged as a technology of choice to tackle problems in speech recognition, computer vision, finance, etc. However, adoption of deep learning in physical domains brings substantial challenges stemming from the correlative nature of deep learning methods compared to the causal, hypothesis driven nature of modern science. We argue that the broad adoption of Bayesian methods incorporating prior knowledge, development of solutions with incorporated physical constraints and parsimonious structural descriptors and generative models, and ultimately adoption of causal models, offers a path forward for fundamental and applied research.
format article
author Rama K. Vasudevan
Maxim Ziatdinov
Lukas Vlcek
Sergei V. Kalinin
author_facet Rama K. Vasudevan
Maxim Ziatdinov
Lukas Vlcek
Sergei V. Kalinin
author_sort Rama K. Vasudevan
title Off-the-shelf deep learning is not enough, and requires parsimony, Bayesianity, and causality
title_short Off-the-shelf deep learning is not enough, and requires parsimony, Bayesianity, and causality
title_full Off-the-shelf deep learning is not enough, and requires parsimony, Bayesianity, and causality
title_fullStr Off-the-shelf deep learning is not enough, and requires parsimony, Bayesianity, and causality
title_full_unstemmed Off-the-shelf deep learning is not enough, and requires parsimony, Bayesianity, and causality
title_sort off-the-shelf deep learning is not enough, and requires parsimony, bayesianity, and causality
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/71fb8b9fb2e94aae97f4e00701a335d9
work_keys_str_mv AT ramakvasudevan offtheshelfdeeplearningisnotenoughandrequiresparsimonybayesianityandcausality
AT maximziatdinov offtheshelfdeeplearningisnotenoughandrequiresparsimonybayesianityandcausality
AT lukasvlcek offtheshelfdeeplearningisnotenoughandrequiresparsimonybayesianityandcausality
AT sergeivkalinin offtheshelfdeeplearningisnotenoughandrequiresparsimonybayesianityandcausality
_version_ 1718392266735747072