Applications of genetic algorithms and neural networks to interatomic potentials

Applications of two modern artificial intelligence (AI) techniques, genetic algorithms (GA) and neural networks (NN) to computer simulations are repor ted. It is shown that the GA are very useful tools for determining the mini mum energy structures of clusters of atoms described by interatomic potenti al functions and generally outperform other optimisation methods for this t ask. A number of applications are given including covalent, and close packe d structures of single or multi-component atomic species. It is also shown that (many body) interatomic potential functions for multi-component system s can be derived by training a specially constructed NN on a variety of str uctural data. (C) 1999 Elsevier Science B.V. All rights reserved.