Sampling, confidence intervals, hypothesis tests and the A/B test you will be asked to design in an interview.
10h · 8 lessons · 1 challenge
Before this: Probability Foundations
An offline metric moved by 0.3%. Is that real? Statistics is how you answer that without embarrassing yourself, and it is the subject ML engineers most often skip and most often get caught on.
The module is built around experiment design, because that is the form the question takes on the job and in the interview: how many samples, what test, what were you assuming, and what would make you wrong.
Run an experiment where you can stop early whenever you like. The game shows you exactly how much significance you manufactured by peeking.
Worth 400 XP.
