Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science

Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science

Artificial neural networks are artificial intelligence computing methods which are inspired by biological neural networks. Here the authors propose a method to design neural networks as sparse scale-free networks, which leads to a reduction in computational time required for training and inference.

Bibliographic Details
Main Authors: Decebal Constantin Mocanu, Elena Mocanu, Peter Stone, Phuong H. Nguyen, Madeleine Gibescu, Antonio Liotta
Format: Article
Language:English
Published: Nature Publishing Group 2018-06-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-018-04316-3
id doaj-50ee9604a82c41788aeb102570ad016f
record_format Article
spelling doaj-50ee9604a82c41788aeb102570ad016f2021-05-11T09:32:03ZengNature Publishing GroupNature Communications2041-17232018-06-019111210.1038/s41467-018-04316-3Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network scienceDecebal Constantin Mocanu0Elena Mocanu1Peter Stone2Phuong H. Nguyen3Madeleine Gibescu4Antonio Liotta5Department of Mathematics and Computer Science, Eindhoven University of TechnologyDepartment of Electrical Engineering, Eindhoven University of TechnologyDepartment of Computer Science, The University of Texas at AustinDepartment of Electrical Engineering, Eindhoven University of TechnologyDepartment of Electrical Engineering, Eindhoven University of TechnologyData Science Centre, University of DerbyArtificial neural networks are artificial intelligence computing methods which are inspired by biological neural networks. Here the authors propose a method to design neural networks as sparse scale-free networks, which leads to a reduction in computational time required for training and inference.https://doi.org/10.1038/s41467-018-04316-3
collection DOAJ
language English
format Article
sources DOAJ
author Decebal Constantin Mocanu
Elena Mocanu
Peter Stone
Phuong H. Nguyen
Madeleine Gibescu
Antonio Liotta
spellingShingle Decebal Constantin Mocanu
Elena Mocanu
Peter Stone
Phuong H. Nguyen
Madeleine Gibescu
Antonio Liotta
Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science
Nature Communications
author_facet Decebal Constantin Mocanu
Elena Mocanu
Peter Stone
Phuong H. Nguyen
Madeleine Gibescu
Antonio Liotta
author_sort Decebal Constantin Mocanu
title Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science
title_short Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science
title_full Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science
title_fullStr Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science
title_full_unstemmed Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science
title_sort scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science
publisher Nature Publishing Group
series Nature Communications
issn 2041-1723
publishDate 2018-06-01
description Artificial neural networks are artificial intelligence computing methods which are inspired by biological neural networks. Here the authors propose a method to design neural networks as sparse scale-free networks, which leads to a reduction in computational time required for training and inference.
url https://doi.org/10.1038/s41467-018-04316-3
work_keys_str_mv AT decebalconstantinmocanu scalabletrainingofartificialneuralnetworkswithadaptivesparseconnectivityinspiredbynetworkscience
AT elenamocanu scalabletrainingofartificialneuralnetworkswithadaptivesparseconnectivityinspiredbynetworkscience
AT peterstone scalabletrainingofartificialneuralnetworkswithadaptivesparseconnectivityinspiredbynetworkscience
AT phuonghnguyen scalabletrainingofartificialneuralnetworkswithadaptivesparseconnectivityinspiredbynetworkscience
AT madeleinegibescu scalabletrainingofartificialneuralnetworkswithadaptivesparseconnectivityinspiredbynetworkscience
AT antonioliotta scalabletrainingofartificialneuralnetworkswithadaptivesparseconnectivityinspiredbynetworkscience
_version_ 1721449641582526464

Similar Items