Knowledge MaxxingLinear regression
Linear regression predicts a continuous value as a weighted sum of input features, learning one coefficient per feature that best fits the training data.
Prerequisite: Regression
What matters
Each coefficient has a direct, interpretable meaning: it estimates how much the predicted value changes when that one feature increases by one unit, holding others constant.
Example
A linear regression model predicting house price might learn a coefficient showing each extra bedroom adds roughly $15,000 to the predicted price.
Common misconception
Linear regression does not assume the relationship is simple just because it is easy to interpret; it specifically assumes the true relationship is close to a straight-line combination of features.
Recall question
What does one coefficient in a linear regression model represent?