https://stevenyuyy.com/notes.html Occasional writing on deep learning, biology, and whatever swims by. en Mon, 06 Jul 2026 00:00:00 GMT https://stevenyuyy.com/writing/ddim-04-sde-to-ode.html https://stevenyuyy.com/writing/ddim-04-sde-to-ode.html Mon, 06 Jul 2026 00:00:00 GMT The finale of the DDIM series. We've removed the Markov shackle (Part 2) and proven the loss is equivalent to DDPM's (Part 3). But de-Markovianizing is only the surface. To… Diffusion Models Generative Models Deep Learning https://stevenyuyy.com/writing/ddim-03-loss-equivalence-proof.html https://stevenyuyy.com/writing/ddim-03-loss-equivalence-proof.html Wed, 01 Jul 2026 00:00:00 GMT Part 3 of the DDIM series. In Part 2 we removed the Markov shackle and derived DDIM's step-skipping sampler under two free-loading requirements. But "it looks like the noising… Diffusion Models Generative Models Deep Learning https://stevenyuyy.com/writing/ddim-02-removing-markov-and-sampling.html https://stevenyuyy.com/writing/ddim-02-removing-markov-and-sampling.html Mon, 29 Jun 2026 00:00:00 GMT Part 2 of the DDIM series. In Part 1 we diagnosed the problem: DDPM's reverse process is shackled to a Markov chain, forcing 1000 one-step-at-a-time denoising steps. Now we… Diffusion Models Generative Models Deep Learning https://stevenyuyy.com/writing/ddim-01-principles-and-ddpm-review.html https://stevenyuyy.com/writing/ddim-01-principles-and-ddpm-review.html Mon, 22 Jun 2026 00:00:00 GMT A hand-holding deep dive into DDIM (Denoising Diffusion Implicit Models). This first part lays out the big picture — what problem DDIM solves and why — then quickly recaps the… Diffusion Models Generative Models Deep Learning https://stevenyuyy.com/writing/ddpm-zero-barrier-04-loss-function.html https://stevenyuyy.com/writing/ddpm-zero-barrier-04-loss-function.html Sun, 21 Jun 2026 00:00:00 GMT The final boss. We've assembled the prerequisites (Part 1), the closed-form noising process (Part 2 — the route), and the denoising posterior (Part 3 — the car). What's left is… Diffusion Models Generative Models Deep Learning https://stevenyuyy.com/writing/ddpm-zero-barrier-03-reverse-process.html https://stevenyuyy.com/writing/ddpm-zero-barrier-03-reverse-process.html Sun, 14 Jun 2026 00:00:00 GMT The third installment. Having built the prerequisites (Part 1) and derived the closed-form forward/noising process (Part 2 — the "route"), we now tackle the reverse/denoising… Diffusion Models Generative Models Deep Learning https://stevenyuyy.com/writing/ddpm-zero-barrier-02-forward-process.html https://stevenyuyy.com/writing/ddpm-zero-barrier-02-forward-process.html Mon, 08 Jun 2026 00:00:00 GMT The second installment of the zero-barrier DDPM series. Here we derive the forward (noising) process in closed form — how to jump from the original image $x0$ to any timestep… Diffusion Models Generative Models Deep Learning https://stevenyuyy.com/writing/ddpm-zero-barrier-01-prerequisites.html https://stevenyuyy.com/writing/ddpm-zero-barrier-01-prerequisites.html Sat, 06 Jun 2026 00:00:00 GMT A beginner-friendly walkthrough of the mathematical prerequisites for reading the Denoising Diffusion Probabilistic Models (DDPM) paper by Jonathan Ho, Ajay Jain, and Pieter… Diffusion Models Generative Models Deep Learning https://stevenyuyy.com/writing/protein-language-models.html https://stevenyuyy.com/writing/protein-language-models.html Fri, 28 Feb 2025 00:00:00 GMT I keep coming back to one strange, wonderful idea: if you treat a protein like a sentence and show a language model enough of them, it starts to understand biology nobody ever… AI Biology Deep Learning