If you can tie in a loss function for a neural network into the quantum excitement state of a quantum system, then presumably, letting the system settle at the

If you can tie in a loss function for a neural network into the quantum excitement state of a quantum system, then presumably, letting the system settle at the energy minimum would be equivalent to a training step, but perhaps much faster.