AlgoPlus//machine learning / diffusion
Read the theory

Diffusion Models

Learn to denoise: destroy an image with noise, then train a network to reverse it.

Phase
Denoise
Noise
Legend
Adding noise
Denoising
AI Tutor Workspace
In a nutshell
A diffusion model learns to generate images by learning to remove noise. In the forward process it takes a real image and adds random noise over many small steps until nothing is left but static. A network is trained to undo one of those steps — to predict and subtract a little noise. To create something new, you start from pure static and run that denoiser over and over (the reverse process), and a coherent image emerges. Condition the denoiser on text and you get tools like Stable Diffusion and DALL·E.
Ready
Press play to begin the cinematic walkthrough.
Learn to clean, then clean from nothing. Practice un-blurring images you deliberately blurred; afterwards, start from pure static and un-blur your way to a brand-new picture.
Key terms
Go deeper in the lesson
Read the full theory, intuition & complexity for Diffusion Models.