🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

1/ Did they fool you into thinking that variational inference is just "the lesser way" of doing Bayesian inference? Please let me argue otherwise. lacerbi.github.io/blog/2024/vi-i… Or just come to play with the interactive VI widget that took me way too long to code up.

I'll be at ICLR this week, hit me up if you want to discuss diffusion models and guidance! We'll be presenting arxiv.org/abs/2410.09945 on thu. 24 at the oral session 1C, then in the afternoon at Hall 2B Poster #143. Come say hi!

10/10 Read the full paper #AABI2025 Proceedings Track! Work by Chengkun Li, Bobby Huggins, Petrus Mikkola, and yours truly. 📜 Paper: arxiv.org/abs/2504.11554 💻 Code: github.com/acerbilab/norm… 📄 Overview: acerbilab.github.io/normalizing-fl… Feedback? Questions? We'd love to hear them!

Daolang Huang will be presenting our work tomorrow on pretrained/amortized inference transformers for all sorts of tasks, go say hi! #AISTATS2025 AISTATS Conference Poster session 1 Place: Hall A-E Time: Sat 3 May 3 p.m. — 6 p.m. Poster number: 23

Website for more info: acerbilab.github.io/amortized-cond…

I want to share my latest (very short) blog post: "Active Learning vs. Data Filtering: Selection vs. Rejection." What is the fundamental difference between active learning and data filtering? Well, obviously, the difference is that: 1/11

Almost exactly two years after we started working on this at Microsoft, I am happy to see the Aurora foundation being published in Nature. Also a nice gift for Silurian's one-year anniversary 🎉 !

New ICML 2025 paper! arxiv.org/abs/2505.20433 Kernel mean embeddings are a super useful tool for representing and comparing distributions through the maximum mean discrepancy (MMD). In this paper, we consider embeddings of distributions using kernel quantiles instead of means.

Videos from the recent UCL workshop on ‘Advances in Post-Bayesian methods’ are now available on YouTube! Check these out if you couldn’t make it: tinyurl.com/AdvPostBayes Jeremias Knoblauch Matias Altamirano Montero

AI benchmarking culture is completely out of control. Tables with dozens of methods, datasets, and bold numbers, trying to answer a question that perhaps no one should be asking anymore.

Composite Goodness-of-fit Tests with Kernels, now out in JMLR! jmlr.org/papers/v26/24-… Test if your distribution comes from ✨any✨ member of a parametric family. Comes in MMD and KSD flavours, and with code. Oscar Key François-Xavier Briol Tamara Fernandez

1/2) It's finally out on Arxiv: Feedback guidance of generative diffusion models! We derived an adaptive guidance methods from first principles that regulate the amount of guidance based on its current state. Complex prompts are highly guided while simplem ones are almost free

Generative models excel at images and text, but tabular data remains a challenge.🤔 We introduce 🐈 TabbyFlow 🐈 - a variational flow matching approach with general exponential families for mixed-type tables. Work with Andrés Guzmán-Cordero & Jan-Willem van de Meent accepted to #ICML2025 🎉 👇 1/n

New blog post! You can train a neural network to *just do things* -- such as *predict the optimum of a function*. But how do you get a big training dataset of "functions with known optima"? Read the blog post to find out! (link 👇)

Blog post: lacerbi.github.io/blog/2025/just…

A great pleasure to crash two Bayesian statistics conferences with a dose of diffusion wisdom — last week in Singapore (bayescomp2025.sg), now in Cambridge (newton.ac.uk/event/rclw03/) — with the two authors of this very nice paper.

Running Distributed Scripts When dealing with running python scripts during distributed training, doing `python myscript(.)py` won't cut it. This is because (especially with multi-node) the same script needs to: 1. Setup PyTorch in such a way that distributed training works

Happens today! 🗓️Tue, July 15 @ 11 AM 📍East Exhibition Hall A-B #E-3512 Unfortunately I was not able to attend, so please DM if you want to chat about hierarchical models, irregular geometries or scalable physical modeling :) Floor Eijkelboom (@ICML 🇨🇦) will present the poster for me on-site

📢 Present your NeurIPS paper in Europe! Join EurIPS 2025 + ELLIS UnConference in Copenhagen for in-person talks, posters, workshops and more. Registration opens soon; save the date: 📅 Dec 2–7, 2025 📍 Copenhagen 🇩🇰 🔗eurips.cc #EurIPS EurIPS Conference

There is an opening for a University Assistant Professor in Machine Learning to be based at Cambridge MLG jobs.cam.ac.uk/job/49361/ Apply!