EVERYTHING AIAI engineering, made visual
Visual edition planned
Phase 08 · Lesson 06BuildPython1.3 h15 lessons in phase

Diffusion Models — DDPM from Scratch

Ho, Jain, Abbeel (2020) gave the field a recipe it could not quit. Destroy the data with noise over a thousand small steps. Train one neural net to predict the noise. Reverse the process at inference. Today every mainstream image, video, 3D, and music model runs on this loop, possibly with flow matching or consistency tricks on top.

Visual edition planned

This lesson isn’t interactive yet.

It is part of the curriculum and will get the same treatment as Phase 1 — a visual cover, hands-on labs, derivations with numeric checks and a quiz. Until then, the original lesson is the best place to read it:

Part of Phase 08Generative AI. Use the previous / next cards below to keep browsing the phase.