Modeling Nuclear Properties with Support Vector Machines

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

Modeling Nuclear Properties with Support Vector Machines

Authors

Description

We have made initial studies of the potential of support vector machines (SVM) for providing statistical models of nuclear systematics with demonstrable predictive power. Using SVM regression and classification procedures, we have created global models of atomic masses, beta-decay halflives, and ground-state spins and parities. These models exhibit performance in both data-fitting and prediction that is comparable to that of the best global models from nuclear phenomenology and microscopic theory, as well as the best statistical models based on multilayer feedforward neural networks.
15 pages; website with latest results added