What Is Supervised Learning
Supervised learning is the most common form of machine learning. You give the model a dataset where every example carries both an input and the correct answer, called the label. The model's job is to learn a function that maps inputs to outputs so it can predict labels for examples it has never seen.
The name comes from the idea of a teacher supervising the student. Each labeled example acts like a graded answer key. During training the model makes a prediction, compares it to the true label, and adjusts itself to reduce the gap.
Supervised tasks split into two families:
- Classification predicts a category, such as spam or not spam
- Regression predicts a continuous number, such as a house price
The central promise is generalization. A model that merely memorizes the training answers is useless. What matters is performance on fresh data drawn from the same distribution.
Supervised learning needs labeled data, which is often expensive to collect because humans must annotate it. Despite that cost, it powers a huge share of practical systems, from medical diagnosis to recommendation.
Key idea
Supervised learning fits a function from inputs to known labels so it can predict answers for unseen data.