Spontaneous representation of numerosity zero in a deep neural network for visual object recognition

Summary: Conceiving “nothing” as a numerical value zero is considered a sophisticated numerical capability that humans share with cognitively advanced animals. We demonstrate that representation of zero spontaneously emerges in a deep learning neural network without any number training. As a signatu...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Khaled Nasr, Andreas Nieder
Formato: article
Lenguaje:EN
Publicado: Elsevier 2021
Materias:
Q
Acceso en línea:https://doaj.org/article/bf057a05c0f849e09117dac907aca694
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:bf057a05c0f849e09117dac907aca694
record_format dspace
spelling oai:doaj.org-article:bf057a05c0f849e09117dac907aca6942021-11-20T05:09:38ZSpontaneous representation of numerosity zero in a deep neural network for visual object recognition2589-004210.1016/j.isci.2021.103301https://doaj.org/article/bf057a05c0f849e09117dac907aca6942021-11-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2589004221012700https://doaj.org/toc/2589-0042Summary: Conceiving “nothing” as a numerical value zero is considered a sophisticated numerical capability that humans share with cognitively advanced animals. We demonstrate that representation of zero spontaneously emerges in a deep learning neural network without any number training. As a signature of numerical quantity representation, and similar to real neurons from animals, numerosity zero network units show maximum activity to empty sets and a gradual decrease in activity with increasing countable numerosities. This indicates that the network spontaneously ordered numerosity zero as the smallest numerical value along the number line. Removal of empty-set network units caused specific deficits in the network's judgment of numerosity zero, thus reflecting these units' functional relevance. These findings suggest that processing visual information is sufficient for a visual number sense that includes zero to emerge and explains why cognitively advanced animals with whom we share a nonverbal number system exhibit rudiments of numerosity zero.Khaled NasrAndreas NiederElsevierarticleBiological sciencesNeuroscienceSensory neuroscienceMachine learningScienceQENiScience, Vol 24, Iss 11, Pp 103301- (2021)
institution DOAJ
collection DOAJ
language EN
topic Biological sciences
Neuroscience
Sensory neuroscience
Machine learning
Science
Q
spellingShingle Biological sciences
Neuroscience
Sensory neuroscience
Machine learning
Science
Q
Khaled Nasr
Andreas Nieder
Spontaneous representation of numerosity zero in a deep neural network for visual object recognition
description Summary: Conceiving “nothing” as a numerical value zero is considered a sophisticated numerical capability that humans share with cognitively advanced animals. We demonstrate that representation of zero spontaneously emerges in a deep learning neural network without any number training. As a signature of numerical quantity representation, and similar to real neurons from animals, numerosity zero network units show maximum activity to empty sets and a gradual decrease in activity with increasing countable numerosities. This indicates that the network spontaneously ordered numerosity zero as the smallest numerical value along the number line. Removal of empty-set network units caused specific deficits in the network's judgment of numerosity zero, thus reflecting these units' functional relevance. These findings suggest that processing visual information is sufficient for a visual number sense that includes zero to emerge and explains why cognitively advanced animals with whom we share a nonverbal number system exhibit rudiments of numerosity zero.
format article
author Khaled Nasr
Andreas Nieder
author_facet Khaled Nasr
Andreas Nieder
author_sort Khaled Nasr
title Spontaneous representation of numerosity zero in a deep neural network for visual object recognition
title_short Spontaneous representation of numerosity zero in a deep neural network for visual object recognition
title_full Spontaneous representation of numerosity zero in a deep neural network for visual object recognition
title_fullStr Spontaneous representation of numerosity zero in a deep neural network for visual object recognition
title_full_unstemmed Spontaneous representation of numerosity zero in a deep neural network for visual object recognition
title_sort spontaneous representation of numerosity zero in a deep neural network for visual object recognition
publisher Elsevier
publishDate 2021
url https://doaj.org/article/bf057a05c0f849e09117dac907aca694
work_keys_str_mv AT khalednasr spontaneousrepresentationofnumerosityzeroinadeepneuralnetworkforvisualobjectrecognition
AT andreasnieder spontaneousrepresentationofnumerosityzeroinadeepneuralnetworkforvisualobjectrecognition
_version_ 1718419564311609344