Introduction
Imagine teaching a child to recognise dogs. You never hand them a rulebook that says "four legs, fur, a tail, barks, between 30 and 90 cm tall…". You just point at many dogs and say "dog", point at cats and say "not a dog", and after a while the child figures out the pattern by themselves. They can even recognise a breed they have never seen before.
Machine learning (ML) is the same idea for computers. Instead of a programmer writing every rule by hand, we show the computer examples, and an algorithm discovers the rules from them.
In this topic you will learn:
- how learning from data differs from ordinary programming;
- the three ingredients that every ML system has: data, a model, and a learning algorithm;
- when ML is a good idea, and when it is not;
- where ML is already part of your everyday life.
No maths is needed yet — this topic is all about the big picture.