Visual lessons and practice
Learn the math behind
machine learning
Work through the mathematics behind machine learning with diagrams and step-by-step examples, then test your understanding with practice questions.
Choose a starting point based on what you already know.
How the dot product compares vectors
A dot product connects coordinates to geometry.
Explore the subjects
From vectors to training a model
Lessons connect through prerequisites, so you can find the background you need before tackling a new concept.
Linear algebra
Represent data with vectors
Vectors, matrices, and the spaces your models learn in.
Explore subject 02Calculus
Calculate how a model changes
Derivatives, gradients, and the chain rule behind backpropagation.
Explore subject 03Probability
Work with uncertain outcomes
Distributions, conditional probability, and information.
Explore subject 04Optimization
Understand parameter updates
From a gradient step to the algorithms that train a model.
Explore subjectHow you learn
Build understanding
through practice
- 01
Make the idea concrete
Follow a worked example alongside a diagram that shows what the equation describes.
- 02
Find out what you understand
Predict an outcome and work through a question. Feedback helps you see where your reasoning holds and where to look again.
- 03
Revisit what you have learned
Return to ideas through continuing practice and review, then build on them in the next lesson.
The Glacius journal
A closer look at the mathematics
Read worked explanations of the questions that come up while learning mathematics and machine learning.