Machine Learning
Algorithms that learn patterns from data instead of being explicitly programmed.
- Fits a model to examples so it can predict on new inputs.
- Contrast: classical programming writes if/else rules by hand.
- Wins when rules are too complex to enumerate.
When to use: When you have many examples but cannot hand-write the rules: vision, language, fraud, recommendation.
Example: A spam filter that learns from millions of labeled emails is supervised ML. Clustering customers by purchase history is unsupervised ML.