Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback

Behavioural tasks often require probability distributions to be inferred about task specific variables. Here, the authors demonstrate that generic neural networks can be trained using a simple error-based learning rule to perform such probabilistic computations efficiently without any need for task...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: A. Emin Orhan, Wei Ji Ma
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2017
Materias:
Q
Acceso en línea:https://doaj.org/article/9eb0080d125246cb88d6ac22181477e7
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:9eb0080d125246cb88d6ac22181477e7
record_format dspace
spelling oai:doaj.org-article:9eb0080d125246cb88d6ac22181477e72021-12-02T14:16:55ZEfficient probabilistic inference in generic neural networks trained with non-probabilistic feedback10.1038/s41467-017-00181-82041-1723https://doaj.org/article/9eb0080d125246cb88d6ac22181477e72017-07-01T00:00:00Zhttps://doi.org/10.1038/s41467-017-00181-8https://doaj.org/toc/2041-1723Behavioural tasks often require probability distributions to be inferred about task specific variables. Here, the authors demonstrate that generic neural networks can be trained using a simple error-based learning rule to perform such probabilistic computations efficiently without any need for task specific operations.A. Emin OrhanWei Ji MaNature PortfolioarticleScienceQENNature Communications, Vol 8, Iss 1, Pp 1-14 (2017)
institution DOAJ
collection DOAJ
language EN
topic Science
Q
spellingShingle Science
Q
A. Emin Orhan
Wei Ji Ma
Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback
description Behavioural tasks often require probability distributions to be inferred about task specific variables. Here, the authors demonstrate that generic neural networks can be trained using a simple error-based learning rule to perform such probabilistic computations efficiently without any need for task specific operations.
format article
author A. Emin Orhan
Wei Ji Ma
author_facet A. Emin Orhan
Wei Ji Ma
author_sort A. Emin Orhan
title Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback
title_short Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback
title_full Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback
title_fullStr Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback
title_full_unstemmed Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback
title_sort efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback
publisher Nature Portfolio
publishDate 2017
url https://doaj.org/article/9eb0080d125246cb88d6ac22181477e7
work_keys_str_mv AT aeminorhan efficientprobabilisticinferenceingenericneuralnetworkstrainedwithnonprobabilisticfeedback
AT weijima efficientprobabilisticinferenceingenericneuralnetworkstrainedwithnonprobabilisticfeedback
_version_ 1718391683553427456