VI — Variational Inference

Fact

A way to replace hard probability math with a faster close-enough answer.

In Plain Words

VI is like picking the fastest grocery line. You skip the coupon wizard and take the close-enough bet.

It helps train probability models and estimate uncertainty faster. The answer is useful, but not exact.

Related Concepts

• Kullback-Leibler Divergence
• Variational Autoencoder
• Generative Model

KL Divergence
Variational Inference uses KL Divergence to measure the gap between its guess and the target.

VAE
VAE training uses Variational Inference as a core way to solve the hard part.

Generative Model
In a Generative Model, Variational Inference often guesses hidden variables after seeing data.