Machine Learning Foundations
A jargon-free introduction to what machine learning is, its three main types, and the data-to-model workflow behind every ML project.
Supervised, unsupervised, reinforcement
Three families of machine learning Almost every machine learning system falls into one of three broad types. They differ in one thing: what kind of examples you give the system, and whether you tell it the right answer. 1. Supervised learning — learning from labelled examples Here every example comes with the correct answer attached, […]
When not to use machine learning
The most valuable ML skill is recognising the problems that are not ML problems Machine learning earns its cost when a rule is genuinely hard to write down and the pattern is genuinely in the data. Plenty of problems meet neither condition, and reaching for a model there buys you complexity, opacity and a maintenance […]
What an accuracy figure is hiding
You will be shown numbers. Here is how to read them You do not need to build models to be useful here. You need to be the person in the room who asks the three questions that separate a real result from a demo. 1. Measured on what? A score on the data the model […]