GLOSSARY TERM

What is Normalizing Flows?

A family of generative models constructing complex distributions from simple ones.
Normalizing flows apply a sequence of invertible, differentiable transformations to a simple base distribution. Because the mappings are bijective, exact log-likelihood evaluation is possible, making them mathematically elegant generative frameworks.

Advanced Generative Modeling

Leverage invertible neural networks directly within the M1 infrastructure.