Computational intelligence applied to the growth of quantum dots

Abstract We apply two computational intelligence techniques, namely, artificial neural network and genetic algorithm to the growth of self-assembled quantum dots. The method relies on an existing database of growth parameters with a resulting quantum dot characteristic to be able to later obtain the growth parameters needed to reach a specific value for such a quantum dot characteristic. The computational techniques were used to associate the growth input parameters with the mean height of the deposited quantum dots. Trends of the quantum dot mean height behavior as a function of growth parameters were correctly predicted and the growth parameters required to minimize the quantum dot mean height were provided.