The interactive film is in the book's Companion

Play it, pause it, turn it and drag it with Mathematics for AI and Machine Learning.

Get the book

Reverse Time: Why We Must Learn the Score

Chapter 21 · Mathematics for AI and Machine Learning · Advanced · 1:26

An interactive live film: reverse diffusion without score blows up into disorder; with Anderson's learned score drift term, random noise contracts uphill against diffusion to reconstruct target data modes.

Narration

Forward diffusion destroys information by turning structured data into random noise. Can we run time backward to create data from noise?

Reversing a stochastic differential equation is fundamentally ill-posed without extra guidance. If you simply reverse the drift, the particles disperse and blow up.

The rest of the narration is in the book's Companion.

More from Chapter 21

Interactive1:35Fokker-Planck: A Random Cloud, a Deterministic DensityChapter 21Preview1:29Detailed Balance: Stationary Is Not the Same as StillChapter 21Preview1:28Forward Diffusion and the Heat Equation: All Distributions Become GaussianChapter 21Preview1:27Guidance Does Not Sample the Sharpened DistributionChapter 21Preview1:50Metastability and the Kramers Escape RateChapter 21Preview1:34Unification: SDEs, Probability Flow ODEs, and Fokker-Planck DynamicsChapter 21Preview1:52Discretization Changes the Stationary DistributionChapter 21Preview1:29Stein Variational Gradient Descent: Pull and PushChapter 21