How machines learn
Regression is a machine learning task where the model predicts a number, such as a price, a temperature, or a delivery time, rather than a category.
Where classification picks a label, regression estimates an amount. How much will this apartment rent for? How many minutes until the bus arrives? How many views will this video get? The output is a point on a scale, and the model learns from past examples with known amounts.
The simplest version draws a straight line through past data and reads predictions off that line. Modern versions use neural networks that can capture curves and interactions between many features. Either way, the model is judged by how far its guesses land from the real values.
The name is a historical accident from 1800s statistics and has nothing to do with going backwards. Do not confuse it with a software regression, which means a feature that used to work has broken. In AI, regression just means predicting numbers.
A food delivery app estimating that your order will arrive in 32 minutes is running a regression model on distance, traffic, and how busy the kitchen is.