Cargando…

Semantic Relatedness Emerges in Deep Convolutional Neural Networks Designed for Object Recognition

Human not only can effortlessly recognize objects, but also characterize object categories into semantic concepts with a nested hierarchical structure. One dominant view is that top-down conceptual guidance is necessary to form such hierarchy. Here we challenged this idea by examining whether deep c...

Descripción completa

Detalles Bibliográficos
Autores principales: Huang, Taicheng, Zhen, Zonglei, Liu, Jia
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7938322/
https://www.ncbi.nlm.nih.gov/pubmed/33692678
http://dx.doi.org/10.3389/fncom.2021.625804
_version_ 1783661578089922560
author Huang, Taicheng
Zhen, Zonglei
Liu, Jia
author_facet Huang, Taicheng
Zhen, Zonglei
Liu, Jia
author_sort Huang, Taicheng
collection PubMed
description Human not only can effortlessly recognize objects, but also characterize object categories into semantic concepts with a nested hierarchical structure. One dominant view is that top-down conceptual guidance is necessary to form such hierarchy. Here we challenged this idea by examining whether deep convolutional neural networks (DCNNs) could learn relations among objects purely based on bottom-up perceptual experience of objects through training for object categorization. Specifically, we explored representational similarity among objects in a typical DCNN (e.g., AlexNet), and found that representations of object categories were organized in a hierarchical fashion, suggesting that the relatedness among objects emerged automatically when learning to recognize them. Critically, the emerged relatedness of objects in the DCNN was highly similar to the WordNet in human, implying that top-down conceptual guidance may not be a prerequisite for human learning the relatedness among objects. In addition, the developmental trajectory of the relatedness among objects during training revealed that the hierarchical structure was constructed in a coarse-to-fine fashion, and evolved into maturity before the establishment of object recognition ability. Finally, the fineness of the relatedness was greatly shaped by the demand of tasks that the DCNN performed, as the higher superordinate level of object classification was, the coarser the hierarchical structure of the relatedness emerged. Taken together, our study provides the first empirical evidence that semantic relatedness of objects emerged as a by-product of object recognition in DCNNs, implying that human may acquire semantic knowledge on objects without explicit top-down conceptual guidance.
format Online
Article
Text
id pubmed-7938322
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-79383222021-03-09 Semantic Relatedness Emerges in Deep Convolutional Neural Networks Designed for Object Recognition Huang, Taicheng Zhen, Zonglei Liu, Jia Front Comput Neurosci Neuroscience Human not only can effortlessly recognize objects, but also characterize object categories into semantic concepts with a nested hierarchical structure. One dominant view is that top-down conceptual guidance is necessary to form such hierarchy. Here we challenged this idea by examining whether deep convolutional neural networks (DCNNs) could learn relations among objects purely based on bottom-up perceptual experience of objects through training for object categorization. Specifically, we explored representational similarity among objects in a typical DCNN (e.g., AlexNet), and found that representations of object categories were organized in a hierarchical fashion, suggesting that the relatedness among objects emerged automatically when learning to recognize them. Critically, the emerged relatedness of objects in the DCNN was highly similar to the WordNet in human, implying that top-down conceptual guidance may not be a prerequisite for human learning the relatedness among objects. In addition, the developmental trajectory of the relatedness among objects during training revealed that the hierarchical structure was constructed in a coarse-to-fine fashion, and evolved into maturity before the establishment of object recognition ability. Finally, the fineness of the relatedness was greatly shaped by the demand of tasks that the DCNN performed, as the higher superordinate level of object classification was, the coarser the hierarchical structure of the relatedness emerged. Taken together, our study provides the first empirical evidence that semantic relatedness of objects emerged as a by-product of object recognition in DCNNs, implying that human may acquire semantic knowledge on objects without explicit top-down conceptual guidance. Frontiers Media S.A. 2021-02-22 /pmc/articles/PMC7938322/ /pubmed/33692678 http://dx.doi.org/10.3389/fncom.2021.625804 Text en Copyright © 2021 Huang, Zhen and Liu. http://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neuroscience
Huang, Taicheng
Zhen, Zonglei
Liu, Jia
Semantic Relatedness Emerges in Deep Convolutional Neural Networks Designed for Object Recognition
title Semantic Relatedness Emerges in Deep Convolutional Neural Networks Designed for Object Recognition
title_full Semantic Relatedness Emerges in Deep Convolutional Neural Networks Designed for Object Recognition
title_fullStr Semantic Relatedness Emerges in Deep Convolutional Neural Networks Designed for Object Recognition
title_full_unstemmed Semantic Relatedness Emerges in Deep Convolutional Neural Networks Designed for Object Recognition
title_short Semantic Relatedness Emerges in Deep Convolutional Neural Networks Designed for Object Recognition
title_sort semantic relatedness emerges in deep convolutional neural networks designed for object recognition
topic Neuroscience
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7938322/
https://www.ncbi.nlm.nih.gov/pubmed/33692678
http://dx.doi.org/10.3389/fncom.2021.625804
work_keys_str_mv AT huangtaicheng semanticrelatednessemergesindeepconvolutionalneuralnetworksdesignedforobjectrecognition
AT zhenzonglei semanticrelatednessemergesindeepconvolutionalneuralnetworksdesignedforobjectrecognition
AT liujia semanticrelatednessemergesindeepconvolutionalneuralnetworksdesignedforobjectrecognition