Skip to content

Glossary: Probability for ML

Entropy

Also written: H(p)

The average surprise of a distribution, −Σ p log p. It measures how unpredictable the distribution is: a fair coin has 1 bit, a certain outcome has 0. It's also the lowest cross-entropy any model can reach.

Where the book teaches it

Related terms

SurpriseCross-entropy (log loss)Bits, natsKL divergence