Top all-time · #9277
Alexia Jolicoeur-Martineau
Personal blog · ajolicoeur.wordpress.comaimath
Alexia Jolicoeur-Martineau's personal blog at ajolicoeur.wordpress.com covers ai and math. Since 2017, 8 different Hacker News readers have linked it 11 times, most often “Score matching with Langevin Sampling: a new contender to GANs”. The last post was in 2024, but the archive is still widely read.
Too few sampled comments (3) to say how often HN recommends it.?
Why HN links it
HN readers link Alexia Jolicoeur-Martineau's blog for GAN and diffusion-model explainers, including a relativistic discriminator post and a 'Deep learning with cats' post with open-source GAN code.
Summarised by Claude from a sample of 3 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Principal Researcher @ Microsoft 🐱💻
Most-linked posts?
- Score matching with Langevin Sampling: a new contender to GANs3 people · 3 mentions · first linked 2020-08-25
- Generating faces of cats using Generative Adversarial Networks3 people · 3 mentions · first linked 2017-07-06
- The relativistic discriminator: a key element missing from standard GAN2 people · 2 mentions · first linked 2018-07-03
What Hacker News says?
Not a lot to ask at all. The go-to person (and one of the pioneers of this type of model) is Alexia: https://twitter.com/jm_alexia The article you want is probably this one: https://ajolicoeur.wordpress.com/the-new-contender-to-gans-s... I would launch into a thorough explanation, but it would likely not be correct in every detail, because it's been around nine months since I was immersing myself in DDPM type models…
Comment linking /the-new-contender-to-gans-score-matching-with-langevin-sampling/
on “Diffusion Models Beat GANs on Image Synthesis”Submitted Score matching with Langevin Sampling: a new contender to GANs as a story · 210 points
HN discussionThe quote is “to lose one is unfortunate, to lose two looks like carelessness.” One person misunderstanding me I can ignore, two in the same way in a short window is definitely a sign I communicated poorly. Why do you believe there is infinite cat behaviour? Why would they evolve that? Even if they did, the point of learning is to reduce a probability distribution from “everything is equally likely, from this cat pa…
Comment linking /cats/
on “Today’s dominant approach to A.I. has not worked out”Submitted Generating faces of cats using Generative Adversarial Networks as a story · 126 points
HN discussionSame guy who wrote this has another post called Deep learning with cats[1] which has a great open source repo[2] with multiple GANs. 1. https://ajolicoeur.wordpress.com/cats/ 2. https://github.com/AlexiaJM/Deep-learning-with-cats
Comment linking /cats/
on “The relativistic discriminator: a key element missing from standard GAN”Submitted The relativistic discriminator: a key element missing from standard GAN as a story · 98 points
HN discussionSubmitted Connections between SVMs, Wasserstein distance and gradient-penalty GANs as a story · 4 points
HN discussionSubmitted On Relativistic F-divergences (AI Research on GANs) as a story · 3 points
HN discussionSubmitted The relativistic discriminator: a key element missing from standard GAN as a story · 3 points
HN discussionSubmitted Adversarial Score Matching and Consistent Sampling (Better Than SOTA GANs) as a story · 2 points
HN discussionSubmitted The new contender to GANs: score matching with Langevin Sampling as a story · 1 points
HN discussion
