Concept guide · Foundations
What is supervised learning?
A slower explanation for when the short interview answer is not enough. Start with the intuition, then connect it to the technical idea and the way an interviewer expects you to reason about it.
ML area
Foundations → Supervised learning
Interview format
Foundations
Question
ML-003
Start with the idea
In simple terms
Supervised learning means learning from examples where the correct answer is already included. The model studies the relationship between the inputs and those answers, then predicts answers for new inputs.
Make it concrete
To predict whether a transaction is fraudulent, each training example can contain transaction features plus a label saying fraudulent or legitimate.
Build the answer
How to reason about it
- 1Name the features available at prediction time.
- 2Name the target or label the model is trained to estimate.
- 3Explain how unseen labelled data is used to evaluate generalisation.
Go one level deeper
The technical explanation
The label is used during training to calculate a loss: the model compares its prediction with the known target and adjusts its parameters. Classification predicts a category or probability, while regression predicts a numeric value. The test of learning is not performance on the labelled examples that shaped the model, but performance on a held-out set that represents future use.
Say this in the interview
Supervised learning estimates a relationship between inputs and a known target. Each training example contains features and a label, such as a transaction and whether it was fraudulent. The model is trained to minimise a loss on those examples, then evaluated on examples whose labels were not used to fit it. Classification and regression are the two common supervised problem types.
Keep preparing