Taming nucleon density distributions with deep neural network

With the datasets of the density distributions calculated by Skyrme density functional theories, we elaborated deep neural networks to generate the density profile and provide a table of related hyperparameters set for similar applications of other structural models. In the process of machine learni...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Zu-Xing Yang, Xiao-Hua Fan, Peng Yin, Wei Zuo
Formato:	article
Lenguaje:	EN
Publicado:	Elsevier 2021
Materias:	Physics QC1-999
Acceso en línea:	https://doaj.org/article/1cd0c4711f174ceeacd82d62fccdcfbf
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:1cd0c4711f174ceeacd82d62fccdcfbf
record_format	dspace
spelling	oai:doaj.org-article:1cd0c4711f174ceeacd82d62fccdcfbf2021-12-04T04:32:20ZTaming nucleon density distributions with deep neural network0370-269310.1016/j.physletb.2021.136650https://doaj.org/article/1cd0c4711f174ceeacd82d62fccdcfbf2021-12-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S0370269321005906https://doaj.org/toc/0370-2693With the datasets of the density distributions calculated by Skyrme density functional theories, we elaborated deep neural networks to generate the density profile and provide a table of related hyperparameters set for similar applications of other structural models. In the process of machine learning with the objective/target functions that normalized mean square error and Kullback–Leibler divergence (cross entropy), there is a turning point showing the transition from the Fermi-like distribution to the realistic Skyrme distribution, while this property is transcended when Pearson χ2 divergence is employed. A training program of about 35 minutes with only about 5%−10% nuclei (200−300) is sufficient to describe the nucleon density distributions of all the nuclear chart within 2% relative error. We obtain similar results employing different datasets calculated by different Skyrme density functional theories. We further investigate the extrapolation properties, which show that an addition of 15 nucleons is acceptable. Based on the results, we propose a mixed dataset approach and a retraining approach in order to go beyond a single physical structure model.Zu-Xing YangXiao-Hua FanPeng YinWei ZuoElsevierarticlePhysicsQC1-999ENPhysics Letters B, Vol 823, Iss , Pp 136650- (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Physics QC1-999
spellingShingle	Physics QC1-999 Zu-Xing Yang Xiao-Hua Fan Peng Yin Wei Zuo Taming nucleon density distributions with deep neural network
description	With the datasets of the density distributions calculated by Skyrme density functional theories, we elaborated deep neural networks to generate the density profile and provide a table of related hyperparameters set for similar applications of other structural models. In the process of machine learning with the objective/target functions that normalized mean square error and Kullback–Leibler divergence (cross entropy), there is a turning point showing the transition from the Fermi-like distribution to the realistic Skyrme distribution, while this property is transcended when Pearson χ2 divergence is employed. A training program of about 35 minutes with only about 5%−10% nuclei (200−300) is sufficient to describe the nucleon density distributions of all the nuclear chart within 2% relative error. We obtain similar results employing different datasets calculated by different Skyrme density functional theories. We further investigate the extrapolation properties, which show that an addition of 15 nucleons is acceptable. Based on the results, we propose a mixed dataset approach and a retraining approach in order to go beyond a single physical structure model.
format	article
author	Zu-Xing Yang Xiao-Hua Fan Peng Yin Wei Zuo
author_facet	Zu-Xing Yang Xiao-Hua Fan Peng Yin Wei Zuo
author_sort	Zu-Xing Yang
title	Taming nucleon density distributions with deep neural network
title_short	Taming nucleon density distributions with deep neural network
title_full	Taming nucleon density distributions with deep neural network
title_fullStr	Taming nucleon density distributions with deep neural network
title_full_unstemmed	Taming nucleon density distributions with deep neural network
title_sort	taming nucleon density distributions with deep neural network
publisher	Elsevier
publishDate	2021
url	https://doaj.org/article/1cd0c4711f174ceeacd82d62fccdcfbf
work_keys_str_mv	AT zuxingyang tamingnucleondensitydistributionswithdeepneuralnetwork AT xiaohuafan tamingnucleondensitydistributionswithdeepneuralnetwork AT pengyin tamingnucleondensitydistributionswithdeepneuralnetwork AT weizuo tamingnucleondensitydistributionswithdeepneuralnetwork
_version_	1718373017047793664

Taming nucleon density distributions with deep neural network

Ejemplares similares