Learning Means What

_Supervised learning_is probably the most common type of machine learning. The goal is simple: using historical data that already has labels (often called the dependent variable), train an algorithm to predict the values of those labels. This training process has the algorithm make a prediction on an input row that is then compared to the true value. The algorithm adjusts internal parameters (often referred to as coefficients) and then repeats this process on the next set of data points. The result of this process is a trained algorithm (often called a model) that you can use to make predictions on new data. There’s a number of different things we’re going to have to do around the process of training and making predictions, like measuring success of trained models before using them in the field, but the fundamental principle is simple: Train on historical data, ensure that it generalizes to data we didn’t train on and then make predictions.