Building a GAN Network
This project will develop a Generative Adersarial Network (GAN), a network that enables generating information after training. The classical 2014 paper is an essential read for developing these networks, although it is written for experts.
We will build our code from an ODE, building on the framework from previous projects.
You will need your backpropagation algorithm code you developed in a previous project for this effort. Allowing for arbitrary inputs, hidden layer neurons, and output neurons will be very valuable in this effort.
You will develop your algorithm for generating sound waveforms.