Predicting regularities in lattice constants of GdFeO3-type perovskites.

A novel idea of employing genetic programming to obtain mathematical expressions representing the dependency of lattice constants (LC) on their atomic parameters is presented in this paper. The results obtained from simulations reveal that only two atomic parameters are sufficient for LC prediction of GdFeO(3)-type perovskites. In addition, an advantage of this approach is that there is no need to save any trained model as in the case of other existing machine-learning based approaches.