Neural network models of nuclear systematics

Gernoth, K. A.; Clark, John W.; Prater, J.S.; Bohr, Henrik

doi:10.1016/0370-2693(93)90738-4

Cited by 70 publications

(68 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The application of artificial neural networks to nuclear physics started in the early 90s with the pioneering work of Clark and collaborators [24][25][26][27], Athanassopoulos and collaborators [28,29] and continues to this day with extensions that include beta-decay systematics that are highly relevant to our understanding of r-process nucleosynthesis [30,31]; for a more recent application of artificial neutral networks to binding-energy systematics see Ref. [32].…”

Section: Introductionmentioning

confidence: 99%

Nuclear mass predictions for the crustal composition of neutron stars: A Bayesian neural network approach

2016

View full text Add to dashboard Cite

Background: Besides their intrinsic nuclear-structure value, nuclear mass models are essential for astrophysical applications, such as r-process nucleosynthesis and neutron-star structure.Purpose: To overcome the intrinsic limitations of existing "state-of-the-art" mass models through a refinement based on a Bayesian Neural Network (BNN) formalism.Methods: A novel BNN approach is implemented with the goal of optimizing mass residuals between theory and experiment.Results: A significant improvement (of about 40%) in the mass predictions of existing models is obtained after BNN refinement. Moreover, these improved results are now accompanied by proper statistical errors. Finally, by constructing a "world average" of these predictions, a mass model is obtained that is used to predict the composition of the outer crust of a neutron star. Conclusions:The power of the Bayesian neural network method has been successfully demonstrated by a systematic improvement in the accuracy of the predictions of nuclear masses. Extension to other nuclear observables is a natural next step that is currently under investigation.

show abstract

Section: Introductionmentioning

confidence: 99%

Nuclear mass predictions for the crustal composition of neutron stars: A Bayesian neural network approach

2016

View full text Add to dashboard Cite

show abstract

“…The results show much better extrapability (i.e. predictive accuracy for new nuclei away from the stable valley) than has been found in earlier neuralnetwork studies 5,6,7 . Moreover, this improvement is achieved with a greatly reduced number of connection weights.…”

Section: Introductionmentioning

confidence: 46%

“…Three different input coding schemes are relevant to our considerations. In the first 5,6,8 , the input layer consists of sixteen "on-off" units having activity levels 0 or 1. Eight units for Z and eight for N serve to encode the proton and neutron numbers in binary and permit the treatment of Z and N values up to 255, which is more than sufficient to cover the interesting physical range of input patterns ν = (Z, N ).…”

Section: Global Models Of Atomic-mass Datamentioning

confidence: 99%

“…The best models from two such investigations are included in Table 1. The network listed in row 6, having architecture (18 + 10 + 10 + 10 + 1)[421], was constructed by Gernoth et al 6 using vanilla back-propagation, with binary encoding of Z and N together with analog encoding of the atomic mass number A and the neutron excess N − Z. The three-layer net (4 + 40 + 1) [245] was provided by Kalman 7 , who employed analog coding of Z and N and auxiliary parity units for these variables.…”

Section: Global Models Of Atomic-mass Datamentioning

confidence: 99%

“…With suitable coding of input and output variables, multilayer feedforward neural networks with pairwise couplings, trained by backpropagation and related algorithms 1,2,3 , are capable of learning from examples in the nuclear database and making predictions for properties of "novel" nuclides outside the training set 4 . Neural network models have been developed for a number of properties, including atomic masses 5,6,7 , neutron separation energies 5 , groundstate spins and parities 8,6 , branching probabilities into different decay channels 9 , and halflives for β − decay 10 . In terms of predictive accuracy, as measured on test nuclei not seen during training, neural-network models can compete with traditional phenomenological and semi-microscopic global models, although the number of adjustable parameters (connection weights) is generally much larger.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Statistical Modeling of Nuclear Systematics

Clark

Mavrommatis

Athanassopoulos

et al. 2001

Fission Dynamics of Atomic Clusters and Nuclei

View full text Add to dashboard Cite

Statistical modeling of data sets by neural-network techniques is offered as an alternative to traditional semiempirical approaches to global modeling of nuclear properties. New results are presented to support the position that such novel techniques can rival conventional theory in predictive power, if not in economy of description. Examples include the statistical inference of atomic masses and β-decay halflives based on the information contained in existing databases. Neural network modeling, as well as other statistical strategies based on new algorithms for artificial intelligence, may prove to be a useful asset in the further exploration of nuclear phenomena far from stability.

show abstract

Explaining the internal behaviour of artificial neural network river flow models

Sudheer

Jain

2004

Hydrological Processes

112

View full text Add to dashboard Cite

A novel method of visualizing and understanding the internal functional behaviour of an artificial neural network (ANN) river flow model is presented. The method hypothesizes that an ANN is able to map a function similar to the flow duration curve while modelling the river flow. A mathematical analysis of the hypothesis is presented, and a case study of an ANN river flow model confirms its significance. The proposed approach is also useful within other models that improve the performance of an ANN. The reasons why these models improve a raw ANN can be clearly understood using this approach. While the field of ANN knowledge-extraction is one that continues to attract considerable interest, it is anticipated that the current approach will initiate further research and make ANNs more useful to the hydrologic community.

show abstract

Neural network models of nuclear systematics

Cited by 70 publications

References 17 publications

Nuclear mass predictions for the crustal composition of neutron stars: A Bayesian neural network approach

Nuclear mass predictions for the crustal composition of neutron stars: A Bayesian neural network approach

Statistical Modeling of Nuclear Systematics

Explaining the internal behaviour of artificial neural network river flow models

Contact Info

Product

Resources

About