Nuclear mass systematics using neural networks

S. Athanassopoulos; E. Mavrommatis; K. A. Gernoth; J. W. Clark

doi:10.1016/j.nuclphysa.2004.08.006

Nuclear mass systematics using neural networks

S. Athanassopoulos, E. Mavrommatis, K. A. Gernoth, J. W. Clark

Research output: Contribution to journal › Article › peer-review

Abstract

New global statistical models of nuclidic (atomic) masses based on multilayered feedforward networks are developed. One goal of such studies is to determine how well the existing data, and only the data, determines the mapping from the proton and neutron numbers to the mass of the nuclear ground state. Another is to provide reliable predictive models that can be used to forecast mass values away from the valley of stability. Our study focuses mainly on the former goal and achieves substantial improvement over previous neural-network models of the mass table by using improved schemes for coding and training. The results suggest that with further development this approach may provide a valuable complement to conventional global models. © 2004 Elsevier B.V. All rights reserved.

Original language	English
Pages (from-to)	222-235
Number of pages	13
Journal	Nuclear Physics A
Volume	743
Issue number	4
DOIs	https://doi.org/10.1016/j.nuclphysa.2004.08.006
Publication status	Published - 1 Nov 2004

Keywords

Binding energies and masses
Neural networks
Statistical modeling

Access to Document

10.1016/j.nuclphysa.2004.08.006

Cite this

@article{43a79be1c16346a39797eecb7c62fae4,

title = "Nuclear mass systematics using neural networks",

abstract = "New global statistical models of nuclidic (atomic) masses based on multilayered feedforward networks are developed. One goal of such studies is to determine how well the existing data, and only the data, determines the mapping from the proton and neutron numbers to the mass of the nuclear ground state. Another is to provide reliable predictive models that can be used to forecast mass values away from the valley of stability. Our study focuses mainly on the former goal and achieves substantial improvement over previous neural-network models of the mass table by using improved schemes for coding and training. The results suggest that with further development this approach may provide a valuable complement to conventional global models. {\textcopyright} 2004 Elsevier B.V. All rights reserved.",

keywords = "Binding energies and masses, Neural networks, Statistical modeling",

author = "S. Athanassopoulos and E. Mavrommatis and Gernoth, {K. A.} and Clark, {J. W.}",

year = "2004",

month = nov,

day = "1",

doi = "10.1016/j.nuclphysa.2004.08.006",

language = "English",

volume = "743",

pages = "222--235",

journal = "Nuclear Physics A",

publisher = "Elsevier BV",

number = "4",

}

TY - JOUR

T1 - Nuclear mass systematics using neural networks

AU - Athanassopoulos, S.

AU - Mavrommatis, E.

AU - Gernoth, K. A.

AU - Clark, J. W.

PY - 2004/11/1

Y1 - 2004/11/1

N2 - New global statistical models of nuclidic (atomic) masses based on multilayered feedforward networks are developed. One goal of such studies is to determine how well the existing data, and only the data, determines the mapping from the proton and neutron numbers to the mass of the nuclear ground state. Another is to provide reliable predictive models that can be used to forecast mass values away from the valley of stability. Our study focuses mainly on the former goal and achieves substantial improvement over previous neural-network models of the mass table by using improved schemes for coding and training. The results suggest that with further development this approach may provide a valuable complement to conventional global models. © 2004 Elsevier B.V. All rights reserved.

AB - New global statistical models of nuclidic (atomic) masses based on multilayered feedforward networks are developed. One goal of such studies is to determine how well the existing data, and only the data, determines the mapping from the proton and neutron numbers to the mass of the nuclear ground state. Another is to provide reliable predictive models that can be used to forecast mass values away from the valley of stability. Our study focuses mainly on the former goal and achieves substantial improvement over previous neural-network models of the mass table by using improved schemes for coding and training. The results suggest that with further development this approach may provide a valuable complement to conventional global models. © 2004 Elsevier B.V. All rights reserved.

KW - Binding energies and masses

KW - Neural networks

KW - Statistical modeling

U2 - 10.1016/j.nuclphysa.2004.08.006

DO - 10.1016/j.nuclphysa.2004.08.006

M3 - Article

VL - 743

SP - 222

EP - 235

JO - Nuclear Physics A

JF - Nuclear Physics A

IS - 4

ER -

Nuclear mass systematics using neural networks

Abstract

Keywords

Access to Document

Fingerprint

Cite this