Demystifying Machine Learning

Summary:

• A model $h(x)$ (or hypothesis) approximates the ‘true’ distribution $f(x)$ that we assume exists.
• The model will learn to estimate $\hat{y}$ from example data $D$ given some input $x$.
• We assume that the example data are drawn from the ‘true’ distribution.
• The goal of the model is usually to have it be as close as possible to the ‘true’ predictor itself.
• An inductive bias constrains the model to be of a specific type.