// ✦ Diffusion, Flow Matching & Generative Models
Denoising Diffusion Probabilistic Models (DDPM)
Learns to turn noise into images step by step.
去噪扩散概率模型 DDPM 2020 Intermediate core
IN PLAIN WORDS
The forward process adds Gaussian noise until pure noise; a network learns to predict that noise. Generation starts from noise and denoises repeatedly. It launched the diffusion era.
Used in: Stable Diffusion, Imagen, DALL·E 2/3, video models
TIMELINE · prerequisites → this → what builds on it
01Original paper
Denoising Diffusion Probabilistic Models
Jonathan Ho, Ajay Jain, Pieter Abbeel · 2020
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Learn first
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
Explainers
Courses
Official docs
04Where the course teaches it
On learning paths:Generative Models & Frontier Research
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
ddpm)