Explain 'Cross-Entropy' for multi-class classification.

Explain 'Cross-Entropy' for multi-class classification.

Approach: Extend binary cross-entropy to multiple classes.

A loss function used when there are more than two classes. It measures the difference between the predicted probability distribution (usually from a Softmax layer) and the actual distribution (one-hot encoded label).

Follow-up: What is Sparse Categorical Cross-Entropy?

Key concepts: softmax, log loss, multi-class.