What is Cross-Entropy loss?

What is Cross-Entropy loss?

Approach: Define the mathematical measure of difference between probability distributions.

A loss function used in classification that measures the performance of a model whose output is a probability value between 0 and 1. It penalizes predictions that are confident and wrong. For binary cases: -[y log(p) + (1-y) log(1-p)].

Follow-up: Why is it preferred over Mean Squared Error for classification?

Key concepts: log loss, softmax, probabilities.