Top all-time · #15861
Yang Song
Personal blog · yang-song.net/blogaidata
Yang Song's personal blog at yang-song.net covers ai and data. Since 2022, 5 different Hacker News readers have linked it 5 times, most often “Generative Modeling by Estimating Gradients of the Data Distribution (2021)”. The last post was in 2021, but the archive is still widely read.
Too few sampled comments (4) to say how often HN recommends it.?
Why HN links it
HN readers link Yang Song's blog on score-based generative models and diffusion models, which explains that diffusion models learn the score function and sample like Langevin dynamics. It is also listed among favourite ML blogs.
Summarised by Claude from a sample of 4 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
The personal website of Yang Song.
Most-linked posts?
- Generative Modeling by Estimating Gradients of the Data Distribution (2021)3 people · 3 mentions · first linked 2023-02-06
What Hacker News says?
Good article, but I feel that it misses an important property of diffusion models that they model the score function (derivative of log prob) [1] and that diffusion sampling is akin to Langevin dynamics [2]. IMO these explain why it's easier to train these models than GANs, because of an easier modeling objective. [1] https://yang-song.net/blog/2021/score/ [2] https://lilianweng.github.io/posts/2021-07-11-diffusion-…
Comment linking /blog/2021/score/
on “Diffusion models from scratch, from a new theoretical perspective”Submitted Generative Modeling by Estimating Gradients of the Data Distribution (2021) as a story · 1 points
HN discussionAlso this blog post https://yang-song.net/blog/2021/score/
Comment linking /blog/2021/score/
on “Tutorial on diffusion models for imaging and vision”Yeah no problem, this is even closer to my area of focus! What do you know about physics and thermodynamics? I'd say a good intro for low background is from Tomczak[0]. He has a book, but the blog posts are nearly identical. He did a post doc with Max Welling (someone you should learn about if you want to get deep, like I was suggesting before). So I'd switch things up slightly. I'd go Intro -> Autoregressive -> Flo…
Comment linking /
on “Stanford A.I. Courses”https://yang-song.net/blog/ https://lilianweng.github.io/
Comment linking /blog/
on “Ask HN: Which are your favourite machine learning blogs?”