An elementary proof of a universal approximation theorem

Chris Monico

In this short note, we give an elementary proof of a universal approximation theorem for neural networks with three hidden layers and increasing, continuous, bounded activation fun…