Concept guide · Foundations
What is machine learning, and how is it different from traditional programming?
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 → What learning means
Interview format
Foundations
Question
ML-001
Start with the idea
In simple terms
Traditional programming asks a person to write the rules. Machine learning asks the computer to infer useful rules from examples, then use those rules on examples it has not seen before.
Make it concrete
For spam detection, you do not write every possible spam rule. You give the model messages labelled spam or not spam, and it learns patterns that help it classify a new message.
Build the answer
How to reason about it
- 1State what the inputs are and what output or decision is required.
- 2Explain that examples are used to estimate a relationship instead of hand-writing every rule.
- 3Finish with generalisation: the model must work on new, representative examples.
Go one level deeper
The technical explanation
The important part is not simply that a model sees data. It learns a mapping from inputs to an output or decision, and that mapping is only valuable if it generalises beyond the training examples. A strong definition therefore includes the data, the target, the learning procedure, and the evaluation setup. If the examples do not represent the real decision, a model can look successful during training while being useless in production.
Common mistake
Saying that the model learns without mentioning data, a target, or generalisation.
Say this in the interview
Traditional programming gives a computer explicit rules and inputs and asks it to produce an output. Machine learning gives the computer examples of inputs and desired outputs, then learns a function that generalises to new inputs. The important qualification is that the learned function is only useful if the training examples, target definition, and evaluation procedure represent the real decision the system will make.
Keep preparing