Convolutional neural networks for vision neuroscience: significance, developments, and outstanding issues
Convolutional Neural Networks (CNN) are a class of machine learning models predominately used in computer vision tasks and can achieve human-like performance through learning from experience. Their striking similarities to the structural and functional principles of the primate visual system allow f...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2023-07-01
|
Series: | Frontiers in Computational Neuroscience |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fncom.2023.1153572/full |
_version_ | 1797786110970036224 |
---|---|
author | Alessia Celeghin Alessio Borriero Davide Orsenigo Matteo Diano Carlos Andrés Méndez Guerrero Alan Perotti Giovanni Petri Marco Tamietto Marco Tamietto |
author_facet | Alessia Celeghin Alessio Borriero Davide Orsenigo Matteo Diano Carlos Andrés Méndez Guerrero Alan Perotti Giovanni Petri Marco Tamietto Marco Tamietto |
author_sort | Alessia Celeghin |
collection | DOAJ |
description | Convolutional Neural Networks (CNN) are a class of machine learning models predominately used in computer vision tasks and can achieve human-like performance through learning from experience. Their striking similarities to the structural and functional principles of the primate visual system allow for comparisons between these artificial networks and their biological counterparts, enabling exploration of how visual functions and neural representations may emerge in the real brain from a limited set of computational principles. After considering the basic features of CNNs, we discuss the opportunities and challenges of endorsing CNNs as in silico models of the primate visual system. Specifically, we highlight several emerging notions about the anatomical and physiological properties of the visual system that still need to be systematically integrated into current CNN models. These tenets include the implementation of parallel processing pathways from the early stages of retinal input and the reconsideration of several assumptions concerning the serial progression of information flow. We suggest design choices and architectural constraints that could facilitate a closer alignment with biology provide causal evidence of the predictive link between the artificial and biological visual systems. Adopting this principled perspective could potentially lead to new research questions and applications of CNNs beyond modeling object recognition. |
first_indexed | 2024-03-13T01:04:22Z |
format | Article |
id | doaj.art-e0bd0fc7971f4caf8bdb8393a62f09f4 |
institution | Directory Open Access Journal |
issn | 1662-5188 |
language | English |
last_indexed | 2024-03-13T01:04:22Z |
publishDate | 2023-07-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Computational Neuroscience |
spelling | doaj.art-e0bd0fc7971f4caf8bdb8393a62f09f42023-07-06T08:17:01ZengFrontiers Media S.A.Frontiers in Computational Neuroscience1662-51882023-07-011710.3389/fncom.2023.11535721153572Convolutional neural networks for vision neuroscience: significance, developments, and outstanding issuesAlessia Celeghin0Alessio Borriero1Davide Orsenigo2Matteo Diano3Carlos Andrés Méndez Guerrero4Alan Perotti5Giovanni Petri6Marco Tamietto7Marco Tamietto8Department of Psychology, University of Torino, Turin, ItalyDepartment of Psychology, University of Torino, Turin, ItalyDepartment of Psychology, University of Torino, Turin, ItalyDepartment of Psychology, University of Torino, Turin, ItalyInstitut des Sciences Cognitives Marc Jeannerod, CNRS, Université de Lyon, Lyon, FranceCENTAI Institute, Turin, ItalyCENTAI Institute, Turin, ItalyDepartment of Psychology, University of Torino, Turin, ItalyDepartment of Medical and Clinical Psychology, and CoRPS–Center of Research on Psychology in Somatic Diseases–Tilburg University, Tilburg, NetherlandsConvolutional Neural Networks (CNN) are a class of machine learning models predominately used in computer vision tasks and can achieve human-like performance through learning from experience. Their striking similarities to the structural and functional principles of the primate visual system allow for comparisons between these artificial networks and their biological counterparts, enabling exploration of how visual functions and neural representations may emerge in the real brain from a limited set of computational principles. After considering the basic features of CNNs, we discuss the opportunities and challenges of endorsing CNNs as in silico models of the primate visual system. Specifically, we highlight several emerging notions about the anatomical and physiological properties of the visual system that still need to be systematically integrated into current CNN models. These tenets include the implementation of parallel processing pathways from the early stages of retinal input and the reconsideration of several assumptions concerning the serial progression of information flow. We suggest design choices and architectural constraints that could facilitate a closer alignment with biology provide causal evidence of the predictive link between the artificial and biological visual systems. Adopting this principled perspective could potentially lead to new research questions and applications of CNNs beyond modeling object recognition.https://www.frontiersin.org/articles/10.3389/fncom.2023.1153572/fullConvolutional Neural Networks (CNN)visual systemventral streamblindsightsuperior colliculuspulvinar |
spellingShingle | Alessia Celeghin Alessio Borriero Davide Orsenigo Matteo Diano Carlos Andrés Méndez Guerrero Alan Perotti Giovanni Petri Marco Tamietto Marco Tamietto Convolutional neural networks for vision neuroscience: significance, developments, and outstanding issues Frontiers in Computational Neuroscience Convolutional Neural Networks (CNN) visual system ventral stream blindsight superior colliculus pulvinar |
title | Convolutional neural networks for vision neuroscience: significance, developments, and outstanding issues |
title_full | Convolutional neural networks for vision neuroscience: significance, developments, and outstanding issues |
title_fullStr | Convolutional neural networks for vision neuroscience: significance, developments, and outstanding issues |
title_full_unstemmed | Convolutional neural networks for vision neuroscience: significance, developments, and outstanding issues |
title_short | Convolutional neural networks for vision neuroscience: significance, developments, and outstanding issues |
title_sort | convolutional neural networks for vision neuroscience significance developments and outstanding issues |
topic | Convolutional Neural Networks (CNN) visual system ventral stream blindsight superior colliculus pulvinar |
url | https://www.frontiersin.org/articles/10.3389/fncom.2023.1153572/full |
work_keys_str_mv | AT alessiaceleghin convolutionalneuralnetworksforvisionneurosciencesignificancedevelopmentsandoutstandingissues AT alessioborriero convolutionalneuralnetworksforvisionneurosciencesignificancedevelopmentsandoutstandingissues AT davideorsenigo convolutionalneuralnetworksforvisionneurosciencesignificancedevelopmentsandoutstandingissues AT matteodiano convolutionalneuralnetworksforvisionneurosciencesignificancedevelopmentsandoutstandingissues AT carlosandresmendezguerrero convolutionalneuralnetworksforvisionneurosciencesignificancedevelopmentsandoutstandingissues AT alanperotti convolutionalneuralnetworksforvisionneurosciencesignificancedevelopmentsandoutstandingissues AT giovannipetri convolutionalneuralnetworksforvisionneurosciencesignificancedevelopmentsandoutstandingissues AT marcotamietto convolutionalneuralnetworksforvisionneurosciencesignificancedevelopmentsandoutstandingissues AT marcotamietto convolutionalneuralnetworksforvisionneurosciencesignificancedevelopmentsandoutstandingissues |