What is machine learning?

Machine learning is a way for computers to learn from data without being explicitly programmed for every possible scenario. Instead of writing rules, you show the computer examples and it figures out the patterns.

[Traditional programming]: You write exact rules. "If the temperature is above 80, turn on the AC."

[Machine learning]: You show examples. "Here are 1000 days of weather data and whether the AC was on. Figure out when to turn it on."

Machine learning systems:

1. ▸[Learn from examples]: They're trained on datasets with thousands or millions of examples
2. ▸[Find patterns]: They identify relationships in the data
3. ▸[Make predictions]: They use those patterns to make decisions about new data

[Supervised learning]: The model learns from labeled examples. You show it pictures of cats and dogs, each labeled, and it learns to tell them apart.

[Unsupervised learning]: The model finds patterns in data without labels. It might discover that customers fall into three groups, even though you didn't tell it to look for groups.

[Reinforcement learning]: The model learns by trial and error, getting rewards for good decisions. This is how AI learns to play games.

• ▸[Email spam filters]: Learn which emails are spam from examples
• ▸[Recommendation systems]: Learn what you like from your past behavior
• ▸[Image recognition]: Learn to identify objects from labeled photos
• ▸[Fraud detection]: Learn to spot suspicious transactions from historical data

Machine learning lets us solve problems that are too complex for traditional programming. When there are too many rules or the rules keep changing, machine learning can adapt automatically.