Finding the energy spectrum of a quantum system is an important but computationally intensive task. The scattering of the system upon an external potential can cause drastic and complicated perturbations to the lowest lying eigenmodes. Even producing a rough estimate of the groundstate, from which a more accurate solution can be derived, is often very difficult.
We propose using neural networks to quickly predict the approximate form of the lowest modes due to a given potential. This first requires we compute the lowest modes for a number of randomly generated potentials, and train our network through supervised machine learning.