distribution

Definition

A specification of which values occur and how often — and, by extension, the spread of the data itself. Most learning theory assumes training and deployment samples come from the same distribution, so the interesting failures are the ones where they do not: distribution shift, out-of-distribution inputs, long tails. Generative modeling goes further and treats the distribution itself as the object being learned and sampled from.

Explainers using this term

• The Mathematics of Diffusion — Generation Seen Through Scores and SDEsGenerative Modeling by Estimating Gradients of the Data Distribution