Models and training

Cross-entropy

stable definition

Definition

A loss that measures how poorly a predicted probability distribution represents the target distribution.

Distinguish it from nearby terms

Cross-entropy is a specific loss over probability distributions. Accuracy counts correct decisions after choosing outputs and does not measure the quality of the full predicted distribution.

Check your understanding

Confident probability assigned to the wrong class produces a larger penalty than a less confident mistake.

Also called

cross-entropy loss

Related terms

Public evidence