What machine learning is
Why this matters
You've used machine-learned models all through Phase 1: every LLM is one. This phase opens the box on the classic kinds. They still run a lot of quiet business machinery. They predict which customers will leave, score which leads are likely to buy, and estimate what a customer will spend over time. They flag fraud and recommend products. They're also the clearest place to learn the ideas that carry over to deep learning and LLMs. Those ideas are training data and generalization (working well on cases the model hasn't seen). They also include overfitting (memorizing instead of learning) and honest testing.
- Contrast writing rules by hand with learning them from examples
- Name the main kinds of learning: supervised, unsupervised, self-supervised and reinforcement
- Judge when machine learning is worth it, and when a rule is better
- See 'learning' as searching for the settings that make the fewest mistakes
Rules in, or rules out
Think about how a child learns what a dog is. Nobody hands them a rulebook: "four legs, fur, a tail." That rulebook would fail anyway, because cats have all of those too. Instead, people point and say "dog" or "not a dog," over and over. The child works out the pattern alone.
Computers can work either way.
In ordinary programming, you write the rules. Say you run an online shop. You might write: "if a customer hasn't ordered in 90 days, flag them as at risk of leaving." Customer data goes in, and answers come out. A customer who stops buying has churned, and the loss of customers is called churn.
In machine learning, you don't write the rule. You give the computer past examples together with the answers. Here that means past customers, and whether each one churned. The computer finds the rule itself:
Traditional programming: rules + data → answers Machine learning: data + answers → rules (a model)
The rule it finds is called a model. A model is a function with some adjustable numbers inside it. Those numbers are its parameters. Think of them as dials, like the dial on an oven. Training means turning the dials until the model's answers match the known answers as closely as possible. Then you use the model on new customers, whose answers you don't know yet.
In plain words: ordinary programming turns rules into answers. Machine learning turns examples with answers into rules.