Supervised Learning
A machine learning approach where models learn from labeled examples that include both inputs and their correct outputs.
Learning With a Teacher
Supervised learning is like learning with an answer key. You show the model an input (like a photo) along with the correct output (like "this is a dog"). After seeing enough examples, the model learns to make accurate predictions on new inputs it hasn't seen before.
The "supervised" part comes from this teaching relationship. Someone has to create those labels - humans looking at images and tagging them, doctors annotating medical scans, or users rating movies. This labeled data is the fuel that powers supervised learning.
Classification vs Regression
Supervised learning handles two main types of problems. Classification assigns inputs to categories - spam or not spam, cat or dog, positive or negative sentiment. Regression predicts continuous values - house prices, temperature forecasts, or how long a delivery will take.
The challenge with supervised learning is getting enough quality labeled data. Labels are expensive and time-consuming to create. They can also contain human biases or errors that the model will learn. But when you have good labeled data, supervised learning is remarkably effective. It's behind most practical AI applications today, from email filtering to medical diagnosis to voice assistants understanding your commands.