Slides

Lecture slides for each day

The deck used in the lecture, as a PDF. Day 1 runs to 62 pages and covers what machine learning is and is not, features and targets, the train and test split, the metrics that match a decision, leakage, and why prediction is not causation. It comes with a separate 101-page primer on the statistics the methods rest on, for anyone who wants that background first. The remaining days are released before each session.