gen-briefly-explain-cross-entropy-and-one-common-misconception-a-180c

Briefly explain cross-entropy and one common misconception about it.

## Cross-Entropy **Cross-entropy** is a measure from information theory that quantifies the difference between two probability distributions. In machine learning, it's commonly used as a loss function by comparing the predicted probability distribution (from your model) against the true distribution (the actual labels). The formula for cross-entropy is: **H(p,q) = -Σ p(x) log(q(x))** where p is the true distribution and q is the predicted distribution. ## Common Misconception **Misconception**: "Cross-entropy and log loss are different things." **Reality**: In binary classification, cross-entropy and log loss are the same thing—just different names for the identical formula. The confusion arises because "log loss" is often used in binary settings while "cross-entropy" is the more general term that applies to both binary and multi-class problems. They're measuring the same concept: how well predicted probabilities match actual outcomes.