Separability and geometry of object manifolds in deep neural networks

Neural activity space or manifold that represents object information changes across the layers of a deep neural network. Here the authors present a theoretical account of the relationship between the geometry of the manifolds and the classification capacity of the neural networks.

Guardado en:

Detalles Bibliográficos
Autores principales:	Uri Cohen, SueYeon Chung, Daniel D. Lee, Haim Sompolinsky
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2020
Materias:	Science Q
Acceso en línea:	https://doaj.org/article/e4533fa30cfe48ebace100730076100b
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:e4533fa30cfe48ebace100730076100b
record_format	dspace
spelling	oai:doaj.org-article:e4533fa30cfe48ebace100730076100b2021-12-02T16:56:34ZSeparability and geometry of object manifolds in deep neural networks10.1038/s41467-020-14578-52041-1723https://doaj.org/article/e4533fa30cfe48ebace100730076100b2020-02-01T00:00:00Zhttps://doi.org/10.1038/s41467-020-14578-5https://doaj.org/toc/2041-1723Neural activity space or manifold that represents object information changes across the layers of a deep neural network. Here the authors present a theoretical account of the relationship between the geometry of the manifolds and the classification capacity of the neural networks.Uri CohenSueYeon ChungDaniel D. LeeHaim SompolinskyNature PortfolioarticleScienceQENNature Communications, Vol 11, Iss 1, Pp 1-13 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	Science Q
spellingShingle	Science Q Uri Cohen SueYeon Chung Daniel D. Lee Haim Sompolinsky Separability and geometry of object manifolds in deep neural networks
description	Neural activity space or manifold that represents object information changes across the layers of a deep neural network. Here the authors present a theoretical account of the relationship between the geometry of the manifolds and the classification capacity of the neural networks.
format	article
author	Uri Cohen SueYeon Chung Daniel D. Lee Haim Sompolinsky
author_facet	Uri Cohen SueYeon Chung Daniel D. Lee Haim Sompolinsky
author_sort	Uri Cohen
title	Separability and geometry of object manifolds in deep neural networks
title_short	Separability and geometry of object manifolds in deep neural networks
title_full	Separability and geometry of object manifolds in deep neural networks
title_fullStr	Separability and geometry of object manifolds in deep neural networks
title_full_unstemmed	Separability and geometry of object manifolds in deep neural networks
title_sort	separability and geometry of object manifolds in deep neural networks
publisher	Nature Portfolio
publishDate	2020
url	https://doaj.org/article/e4533fa30cfe48ebace100730076100b
work_keys_str_mv	AT uricohen separabilityandgeometryofobjectmanifoldsindeepneuralnetworks AT sueyeonchung separabilityandgeometryofobjectmanifoldsindeepneuralnetworks AT danieldlee separabilityandgeometryofobjectmanifoldsindeepneuralnetworks AT haimsompolinsky separabilityandgeometryofobjectmanifoldsindeepneuralnetworks
_version_	1718382831738028032

Separability and geometry of object manifolds in deep neural networks

Ejemplares similares