Diffusion models for images

Level 7: Generative vision

Coming soon

This page is being written. See the full roadmap for the whole syllabus.

What this chapter will cover

• The forward and reverse processes
• the noise-prediction objective
• U-Net denoisers with timestep embeddings
• DDPM and DDIM sampling
• classifier-free guidance
• latent diffusion and Stable Diffusion's architecture