Caesar AI Atlas

Relative Squared Error

Caesar AI Atlas Definition

Relative squared error is a regression evaluation metric that compares a model's total squared error with the squared error of a baseline that predicts the average target value. It emphasizes larger errors more heavily than relative absolute error.

Other Definitions

Relative Squared Error Source

In regression , an evaluation metric that is the sum of all squared absolute errors divided by the sum of squared distances between correct label values and the average of all correct label values.

Related Terms