Abstract

Over the past several years big data methods, including but not limited to use of deep convolutional neural networks, have been very successful in computer science applications and there is increasing effort to apply big data or machine learning methods to problems in physical science and engineering. Conversely we are seeing that problems from physical science are influencing machine learning research done in computer science environments. In the talk I will provide a survey of recently developed methods for the construction of effective interatomic potentials and force fields for atomistic modelling of molecular and condensed phase systems. I will also show how this work is influencing developments in machine learning through the concept of deep neural networks that are invariant or covariant (equivariant) with respect to groups of discrete or continuous transformations. In the case of atomistic systems these transformations are the spatial point group of translations, rotations and reflections and the permutational symmetry group associated with the labeling of chemically identical atoms.