01Available to explore
Algebra, functions & graphs
Explain how inputs become outputs through a rule.
Variables, equations, graphs, powers and logarithms. Recommended first subject.
Open algebra & functions01 · Math foundations
Make the mathematics
mean something.
Drag, predict and code your way through the maths behind machine learning. Every idea starts with a question you can answer by playing.
Why and how
Math for machine learning
See which ideas each algorithm needs and why. Open the Math-to-ML map, or go straight to a bridge lesson: gradients, the normal curve, least squares, likelihood.
Recommended sequence
Choose a foundation.
Six subjects. Each has playgrounds to drag, Python to run and a short assessment when you want to test yourself.
02Available to explore
Linear algebra
Trace several inputs through vectors and matrices.
Vectors, dot products, matrices, transformations and eigenvector intuition.
Open linear algebra03Available to explore
Calculus
Explain how a small change affects a prediction or loss.
Local change, derivatives, the chain rule and gradients. Review functions first if needed.
Open calculus04Available to explore
Probability
Reason about uncertain outcomes and conditional evidence.
Events, conditioning, Bayes and distributions. Helpful before classification.
Open probability05Available to explore
Statistics
Describe data and recognize what a sample can justify.
Center, spread, relationships and sampling, with limits on causal claims.
Open statistics06Available to explore
Optimization
Explain how parameter updates can reduce a defined loss.
Objectives, gradients, step size, stability and regularization. Review calculus first.
Open optimization07Available to explore
Math bridges to ML
Use distance, expectation and likelihood the way ML does.
Norms, expected value and variance, maximum likelihood. Story-led lessons for the ideas other modules lean on.
Open math bridgesBring the ideas into code.
Ready to connect these concepts to arrays, data tables and fitted workflows?
Explore Python →