Machine learning
Also known as: ML
Machine learning (ML) is a way of building software where, instead of you writing the rules, the program figures out the rules itself from examples.
You show it thousands of photos labeled “cat” or “not cat”, and it adjusts internal numbers until it can tell the difference on new photos. You never wrote “if pointed ears and whiskes then cat” — it discovered the pattern.
- Supervised — examples come with the right answer (labeled data).
- Unsupervised — no labels; the model finds structure on its own.
- Reinforcement — the model learns by trial, error, and reward.
Why it matters to you: every modern LLM, image model, and voice assistant is a machine-learning system. "Training" a model = running machine learning on a giant pile of examples.