Silicon potentials investigated using density functional theory fitted neural networks

We present a method for fitting neural networks to geometric and energetic data sets. We then apply this method by fitting a neural network to a set of data generated using the local density approximation for systems composed entirely of silicon. In order to generate atomic potential energy data, we...

Full description

Bibliographic Details

Main Authors:	E. Sanville, Ajeevsing Bholoa, Roger Smith, Steven Kenny
Format:	Default Article
Published:	2008
Subjects:	Other mathematical sciences not elsewhere classified Condensed matter physics not elsewhere classified untagged Mathematical Sciences not elsewhere classified Condensed Matter Physics
Online Access:	https://hdl.handle.net/2134/15407
Tags:	Add Tag No Tags, Be the first to tag this record!