Predictive learning as a network mechanism for extracting low-dimensional latent space representations

Neural networks trained using predictive models generate representations that recover the underlying low-dimensional latent structure in the data. Here, the authors demonstrate that a network trained on a spatial navigation task generates place-related neural activations similar to those observed in...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Stefano Recanatesi, Matthew Farrell, Guillaume Lajoie, Sophie Deneve, Mattia Rigotti, Eric Shea-Brown
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
Q
Acceso en línea:https://doaj.org/article/411ac526db244399bdd114924abfd060
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:411ac526db244399bdd114924abfd060
record_format dspace
spelling oai:doaj.org-article:411ac526db244399bdd114924abfd0602021-12-02T13:32:57ZPredictive learning as a network mechanism for extracting low-dimensional latent space representations10.1038/s41467-021-21696-12041-1723https://doaj.org/article/411ac526db244399bdd114924abfd0602021-03-01T00:00:00Zhttps://doi.org/10.1038/s41467-021-21696-1https://doaj.org/toc/2041-1723Neural networks trained using predictive models generate representations that recover the underlying low-dimensional latent structure in the data. Here, the authors demonstrate that a network trained on a spatial navigation task generates place-related neural activations similar to those observed in the hippocampus and show that these are related to the latent structure.Stefano RecanatesiMatthew FarrellGuillaume LajoieSophie DeneveMattia RigottiEric Shea-BrownNature PortfolioarticleScienceQENNature Communications, Vol 12, Iss 1, Pp 1-13 (2021)
institution DOAJ
collection DOAJ
language EN
topic Science
Q
spellingShingle Science
Q
Stefano Recanatesi
Matthew Farrell
Guillaume Lajoie
Sophie Deneve
Mattia Rigotti
Eric Shea-Brown
Predictive learning as a network mechanism for extracting low-dimensional latent space representations
description Neural networks trained using predictive models generate representations that recover the underlying low-dimensional latent structure in the data. Here, the authors demonstrate that a network trained on a spatial navigation task generates place-related neural activations similar to those observed in the hippocampus and show that these are related to the latent structure.
format article
author Stefano Recanatesi
Matthew Farrell
Guillaume Lajoie
Sophie Deneve
Mattia Rigotti
Eric Shea-Brown
author_facet Stefano Recanatesi
Matthew Farrell
Guillaume Lajoie
Sophie Deneve
Mattia Rigotti
Eric Shea-Brown
author_sort Stefano Recanatesi
title Predictive learning as a network mechanism for extracting low-dimensional latent space representations
title_short Predictive learning as a network mechanism for extracting low-dimensional latent space representations
title_full Predictive learning as a network mechanism for extracting low-dimensional latent space representations
title_fullStr Predictive learning as a network mechanism for extracting low-dimensional latent space representations
title_full_unstemmed Predictive learning as a network mechanism for extracting low-dimensional latent space representations
title_sort predictive learning as a network mechanism for extracting low-dimensional latent space representations
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/411ac526db244399bdd114924abfd060
work_keys_str_mv AT stefanorecanatesi predictivelearningasanetworkmechanismforextractinglowdimensionallatentspacerepresentations
AT matthewfarrell predictivelearningasanetworkmechanismforextractinglowdimensionallatentspacerepresentations
AT guillaumelajoie predictivelearningasanetworkmechanismforextractinglowdimensionallatentspacerepresentations
AT sophiedeneve predictivelearningasanetworkmechanismforextractinglowdimensionallatentspacerepresentations
AT mattiarigotti predictivelearningasanetworkmechanismforextractinglowdimensionallatentspacerepresentations
AT ericsheabrown predictivelearningasanetworkmechanismforextractinglowdimensionallatentspacerepresentations
_version_ 1718392848277045248