This short tutorial covers the basics of diffusion models, a simple yet expressive approach to generative modeling. They've been behind a recent string of impressive results, including OpenAI's DALL-E 2, Google's Imagen, and Stable Diffusion.
Errata:
At 12:39, parentheses are missing around the difference: \epsilon(x, t, y) - \epsilon(x, t, \empty). See [ Ссылка ] for corrected version.
Timestamps:
0:00 - Intro
1:07 - Forward process
3:07 - Posterior of forward process
4:16 - Reverse process
5:34 - Variational lower bound
9:26 - Reduced variance objective
10:27 - Reverse step implementation
11:38 - Conditional generation
13:45 - Comparison with other deep generative models
14:34 - Connection to score matching models
Special thanks to Jonathan Ho and Elmira Amirloo for feedback on this video.
Papers:
Feller, 1949: On the Theory of Stochastic Processes, with Particular Reference to Applications ([ Ссылка ])
Sohl-Dickstein et al., 2015: Deep Unsupervised Learning using Nonequilibrium Thermodynamics ([ Ссылка ])
Ho et al., 2020: Denoising Diffusion Probabilistic Models ([ Ссылка ])
Song & Ermon, 2019: Generative Modeling by Estimating Gradients of the Data Distribution ([ Ссылка ])
Dhariwal & Nichol, 2021: Diffusion Models Beat GANs on Image Synthesis ([ Ссылка ])
Nichol et al., 2021: GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models ([ Ссылка ])
Saharia et al., 2021: Palette: Image-to-Image Diffusion Models ([ Ссылка ])
Ramesh et al, 2022: Hierarchical Text-Conditional Image Generation with CLIP Latents ([ Ссылка ])
Saharia et al., 2022: Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding ([ Ссылка ])
Song et al., 2021: Denoising Diffusion Implicit Models ([ Ссылка ])
Nichol & Dhariwal, 2021: Improved Denoising Diffusion Probabilistic Models ([ Ссылка ])
Kingma et al., 2021: Variational Diffusion Models ([ Ссылка ])
Song et al., 2021: Score-Based Generative Modeling through Stochastic Differential Equations ([ Ссылка ])
Links:
YouTube: [ Ссылка ]
Twitter: [ Ссылка ]
Homepage: [ Ссылка ]
If you'd like to help support the channel (completely optional), you can donate a cup of coffee via the following:
Venmo: [ Ссылка ]
PayPal: [ Ссылка ]
