In this subchapter, we introduce the architecture of single hidden layer neural networks and discuss the advantages of hidden layers. In addition, we present some typical (non-linear) activation functions.
