Maths · Module 1Module material

Functions, Logarithms and Notation interview questions

15 questions interviewers ask on this module’s topics, from first principles to the follow-ups. Each comes with a model answer.

What makes a rule a function?
How can you tell a linear, a quadratic and an exponential relationship apart from a table?
What does f(x − 3) do to a graph, and how do you remember the direction?
Your straight line passes through the middle of the data. How do you check that it is the right shape?
Why is e the base used in machine learning?
State the product rule for logarithms and the most common mistake with it.
Why do we work with log-probabilities instead of probabilities?
What are odds and log-odds, and why does logistic regression use the log-odds?
An email scores 5.872 in spam-minus-ham log-probability. What is the probability it is spam?
You log-transform a column and get -inf. What happened and what do you do?
What do ŷ, x̄ and w* mean?
Write ŷᵢ = Σⱼ wⱼ xᵢⱼ in NumPy, and explain the index shift.
What is the difference between min and argmin?
scikit-learn's Ridge and a paper both use a regularisation strength. Can you copy the paper's value?
Ridge's penalty is λ Σⱼ wⱼ². On the shop data, promotion's weight is 99.48% of that sum. Is that a problem?