Neural Nets As Universal Approximators

This week we’re starting with Multilayer Perceptrons (MLPs), the most basic form of neural networks. We’ll go over their architecture, activation functions, and see how they serve as universal function approximators.