Glossary — Fundamentals
Machine learning
Machine learning is a way of getting a computer to solve a task without anybody writing the rules for it — it derives the rules itself, from examples.
2 tokensMachinelearning
Simply put#
An ordinary program follows rules a programmer wrote down. With machine learning you give the computer a pile of examples and it works the rules out itself.
Example#
Nobody writes a list saying "these words mean spam". Instead the filter is shown hundreds of thousands of messages labelled spam or not, and learns the patterns on its own.
Why it matters to you#
This is where AI's most characteristic property comes from: it is only as good as its examples. Biased data gives a biased result, and nobody ever wrote that bias down as a rule.