Nonlinear reconfiguration of network edges, topology and information content during an artificial learning task

Abstract Here, we combine network neuroscience and machine learning to reveal connections between the brain’s network structure and the emerging network structure of an artificial neural network. Specifically, we train a shallow, feedforward neural network to classify hand-written digits and then us...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	James M. Shine, Mike Li, Oluwasanmi Koyejo, Ben Fulcher, Joseph T. Lizier
Formato:	article
Lenguaje:	EN
Publicado:	SpringerOpen 2021
Materias:	Network Integration Low-dimensional Information Pixels Reconfiguration Computer applications to medicine. Medical informatics R858-859.7 Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/56535ee1d8f24f75addbbf7d10541b34
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:56535ee1d8f24f75addbbf7d10541b34
record_format	dspace
spelling	oai:doaj.org-article:56535ee1d8f24f75addbbf7d10541b342021-12-05T12:06:57ZNonlinear reconfiguration of network edges, topology and information content during an artificial learning task10.1186/s40708-021-00147-z2198-40182198-4026https://doaj.org/article/56535ee1d8f24f75addbbf7d10541b342021-12-01T00:00:00Zhttps://doi.org/10.1186/s40708-021-00147-zhttps://doaj.org/toc/2198-4018https://doaj.org/toc/2198-4026Abstract Here, we combine network neuroscience and machine learning to reveal connections between the brain’s network structure and the emerging network structure of an artificial neural network. Specifically, we train a shallow, feedforward neural network to classify hand-written digits and then used a combination of systems neuroscience and information-theoretic tools to perform ‘virtual brain analytics’ on the resultant edge weights and activity patterns of each node. We identify three distinct phases of network reconfiguration across learning, each of which are characterized by unique topological and information-theoretic signatures. Each phase involves aligning the connections of the neural network with patterns of information contained in the input dataset or preceding layers (as relevant). We also observe a process of low-dimensional category separation in the network as a function of learning. Our results offer a systems-level perspective of how artificial neural networks function—in terms of multi-stage reorganization of edge weights and activity patterns to effectively exploit the information content of input data during edge-weight training—while simultaneously enriching our understanding of the methods used by systems neuroscience.James M. ShineMike LiOluwasanmi KoyejoBen FulcherJoseph T. LizierSpringerOpenarticleNetworkIntegrationLow-dimensionalInformationPixelsReconfigurationComputer applications to medicine. Medical informaticsR858-859.7Computer softwareQA76.75-76.765ENBrain Informatics, Vol 8, Iss 1, Pp 1-15 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Network Integration Low-dimensional Information Pixels Reconfiguration Computer applications to medicine. Medical informatics R858-859.7 Computer software QA76.75-76.765
spellingShingle	Network Integration Low-dimensional Information Pixels Reconfiguration Computer applications to medicine. Medical informatics R858-859.7 Computer software QA76.75-76.765 James M. Shine Mike Li Oluwasanmi Koyejo Ben Fulcher Joseph T. Lizier Nonlinear reconfiguration of network edges, topology and information content during an artificial learning task
description	Abstract Here, we combine network neuroscience and machine learning to reveal connections between the brain’s network structure and the emerging network structure of an artificial neural network. Specifically, we train a shallow, feedforward neural network to classify hand-written digits and then used a combination of systems neuroscience and information-theoretic tools to perform ‘virtual brain analytics’ on the resultant edge weights and activity patterns of each node. We identify three distinct phases of network reconfiguration across learning, each of which are characterized by unique topological and information-theoretic signatures. Each phase involves aligning the connections of the neural network with patterns of information contained in the input dataset or preceding layers (as relevant). We also observe a process of low-dimensional category separation in the network as a function of learning. Our results offer a systems-level perspective of how artificial neural networks function—in terms of multi-stage reorganization of edge weights and activity patterns to effectively exploit the information content of input data during edge-weight training—while simultaneously enriching our understanding of the methods used by systems neuroscience.
format	article
author	James M. Shine Mike Li Oluwasanmi Koyejo Ben Fulcher Joseph T. Lizier
author_facet	James M. Shine Mike Li Oluwasanmi Koyejo Ben Fulcher Joseph T. Lizier
author_sort	James M. Shine
title	Nonlinear reconfiguration of network edges, topology and information content during an artificial learning task
title_short	Nonlinear reconfiguration of network edges, topology and information content during an artificial learning task
title_full	Nonlinear reconfiguration of network edges, topology and information content during an artificial learning task
title_fullStr	Nonlinear reconfiguration of network edges, topology and information content during an artificial learning task
title_full_unstemmed	Nonlinear reconfiguration of network edges, topology and information content during an artificial learning task
title_sort	nonlinear reconfiguration of network edges, topology and information content during an artificial learning task
publisher	SpringerOpen
publishDate	2021
url	https://doaj.org/article/56535ee1d8f24f75addbbf7d10541b34
work_keys_str_mv	AT jamesmshine nonlinearreconfigurationofnetworkedgestopologyandinformationcontentduringanartificiallearningtask AT mikeli nonlinearreconfigurationofnetworkedgestopologyandinformationcontentduringanartificiallearningtask AT oluwasanmikoyejo nonlinearreconfigurationofnetworkedgestopologyandinformationcontentduringanartificiallearningtask AT benfulcher nonlinearreconfigurationofnetworkedgestopologyandinformationcontentduringanartificiallearningtask AT josephtlizier nonlinearreconfigurationofnetworkedgestopologyandinformationcontentduringanartificiallearningtask
_version_	1718372254363942912

Nonlinear reconfiguration of network edges, topology and information content during an artificial learning task

Ejemplares similares