Evolutionary approach for determining first-principles hamiltonians

The use of genetic algorithms to distil reliable model hamiltonian parameter from a database of first-principles calculations was discussed. The selection of leading parameters based on a genetic algorithm is general and easily applied to construct any other type of complex model hamiltonian from direct quantum-mechanical results. It was shown that the genetic algorithm can be used to select the best components of model hamiltonians from a vast set of possibilities using a handful of first-principles calculations as a guide. The unknown ground states for two chemically distinct systems with the accuracy of the first-principles hamiltonian were successfully predicted.