Normalizing Flow

Fact

A generative model that turns simple randomness into complex data using reversible steps.

In Plain Words

Normalizing Flow is like a perfect Play-Doh machine. It squishes a plain blob into a unicorn, then can unsquish it back.

It can make new samples and score exact odds. Its tricks must stay reversible, so the design has tight rules.

Related Concepts

• Generative Model
• Variational Inference
• Diffusion Model

Generative Model
Normalizing Flow is a Generative Model that can compute probabilities directly.

VI
Normalizing Flow can make the posterior guess in VI more flexible.

Diffusion
Normalizing Flow and Diffusion both generate samples, but they follow different paths.