Mathematics for Machine Learnings
Mathematics for machine learning is the linear algebra, calculus and probability that models are built on. Vectors and matrices represent data, derivatives guide gradient descent, and probability and statistics describe uncertainty and distributions. This unit sits in the Maths and Programming stage and makes later algorithms understandable rather than mysterious.
Stage: Maths and Programming · Part of the Machine Learning roadmap · Last reviewed 19 November 2026
What to learn in Mathematics
- Compute a dot product and a matrix multiplication and explain what they mean for data
- Take a partial derivative and describe how gradient descent uses it
- Explain the roles of the learning rate and the loss surface
- Apply Bayes' rule to a simple diagnostic-test example
- Interpret mean, variance, the normal distribution and a confidence interval
Mathematics interview questions
What is gradient descent?
Gradient descent is an optimisation method that repeatedly moves model parameters a small step in the direction opposite to the gradient of the loss, which reduces the loss. The learning rate sets the step size. Too large overshoots, too small is slow.
What is the difference between variance and standard deviation?
Variance is the average squared distance of values from the mean, so its units are squared. Standard deviation is its square root, returning to the original units and making spread easier to interpret, for example in a normal distribution.
Where Mathematics fits in the Machine Learning roadmap
Mathematics belongs to the Maths and Programming stage (beginner level), which takes about 4 hours. The toolkit.
Other units in this stage
Back to the full Machine Learning roadmap