Diffusion & Image Generation · Basics
Diffusion & Image Generation — Basics
20 practice questions on Diffusion & Image Generation. Every question is written from a specific moment in a real lecture, and after you answer it links to that exact timestamp so you can check it yourself.
What this pack asks
The free questions in this pack. Answer choices and explanations appear as you play.
In training a denoising model for image generation, how are the input-output pairs built from training images?
Answer this one →In most modern diffusion models, what does the network output at each denoising step?
Answer this one →To classify an image with CLIP given a fixed list of candidate labels, what procedure is used?
Answer this one →
Written from lectures by MIT OpenCourseWare. Not affiliated with or endorsed by any university, channel, creator or certification program.