Diffusion Models Beat GANs on Image Synthesis - ML Coding Series - Part 2

Diffusion Models Beat GANs on Image Synthesis - ML Coding Series - Part 2

Aleksa Gordić - The AI Epiphany via YouTube Direct link

Core of the sampling logic

11 of 14

11 of 14

Core of the sampling logic

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Diffusion Models Beat GANs on Image Synthesis - ML Coding Series - Part 2

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Intro
  2. 2 Paper overview part - U-Net architecture improvements
  3. 3 Classifier guidance explained
  4. 4 Intuition behind classifier guidance
  5. 5 Scaling classifier guidance
  6. 6 Diversity vs quality tradeoff and future work
  7. 7 Coding part - training a noise-aware classifier
  8. 8 Main training loop
  9. 9 Visualizing timestep conditioning
  10. 10 Sampling using classifier guidance
  11. 11 Core of the sampling logic
  12. 12 Shifting the mean - classifier guidance
  13. 13 Minor bug in their code and my GitHub issue
  14. 14 Outro

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.