Jan-Willem van de Meent

At #ICML2025 this week?

Come check out our work on controlled generation, simulation-free latent SDEs, and risk monitoring in test-time adaptation, and chat with the awesome students that made it happen!

#SDE #Diffusion #FlowMatching #TTA #UncertaintyQuantification

Are you interested in improving the #interpretability, #robustness and #safety of current AI systems with #causality and #RL?

Apply to our PhD position in Amsterdam 🚲🌷🇳🇱

Deadline: June 15

🐳 Controlled Generation with Equivariant Variational Flow Matching

By @eijkelboomfloor.bsky.social , @zmheiko.bsky.social ,
@sharvaree.bsky.social , @erikjbekkers.bsky.social ,
@wellingmax.bsky.social , @canaesseth.bsky.social *, @jwvdm.bsky.social *

📜 As above - paper TBA soon 😉

🧵8 / 8

🌊 Exponential Family Variational Flow Matching for Tabular Data Generation

By Andrés Guzmán-Cordero*, @eijkelboomfloor.bsky.social *,
@jwvdm.bsky.social

📜 This paper will be shared soon - keep your eyes open! 🤩

🧵7 / 8

🐸 Erwin: A Tree-based Hierarchical Transformer for Large-scale Physical Systems

By @maxxxzdn.bsky.social , @jwvdm.bsky.social , @wellingmax.bsky.social

📜 arxiv.org/abs/2502.17019

🧵2 / 8

Exciting news: AMLab is happy to have 7 papers accepted at #ICML2025! 🎉

See the thread below for the full list 📝 and meet us in Vancouver to discuss them further! 🇨🇦

🧵1 / 8

🥇 3 Best Paper Awards for AMLab members!

1. Towards Variational Flow Matching on General Geometries by @olgatticus.bsky.social et al.

2. Generative Uncertainty in Diffusion Models by @metodjazbec.bsky.social et al.

3. SDE Matching by @gbarto.bsky.social et al.

Congrats to everyone! 🔥

Nice work! We were ourselves also wondering why few approaches seem to consider conditional entropy as a function of time (beyond the time-warping in the work of @sedielem.bsky.social, as you note). Great to see that this can aid generation in other settings as well.

Postdoc position: Computational methods for neurodegeneration research - VIB About the labs The Verstreken Lab and the Gonçalves Lab are based in Leuven, Belgium and part of the

Exciting project in Parkinson's disease! Patrik Verstreken's lab (verstrekenlab.sites.vib.be) and my lab are looking for a postdoc to develop and apply ML methods to the analysis of exciting brain-on-chip data. Please spread the word!

More details at: jobs.vib.be/j/110908/pos...

Our colleagues at Climate AI Nordics are seeking collabroators in Iceland, the Faroe Islands, Greenland, and Åland 📣

Please reshare this with your networks if you have colleagues who might be interested in joining this team 🌍

CmdStanR: v0.9.0 Release Hello all, I’m excited to announce that we have now released CmdStanR v0.9.0! This is a minor release, wrapping up several months of bugfixes, quality-of-life improvements, and updates to documentatio...

CmdStanR v 0.9.0 released! According to Andrew Johnson, this should make Windows installation easier plus many other improvements and bug fixes discourse.mc-stan.org/t/cmdstanr-v...

#Bayesian #MCMC

A class photo in front of the AIMS building in Muizenberg, Cape Town.

A selfie with students at AIMS in Muizenberg, Cape Town.

Just returned from an inspiring three weeks at AIMS (aims.ac.za) in Cape Town, where I taught Generative AI with @eijkelboomfloor.bsky.social.

I was really impressed with these students, who are engaged, very hard working, and asked some of the best questions that I have ever had in class!

Excellent work by @maxxxzdn
that uses ideas from computational physics to design scalable transformers!

Research Scientist, Generative Media London, UK

We are hiring on the Generative Media team in London: boards.greenhouse.io/deepmind/job...

We work on Imagen, Veo, Lyria and all that good stuff. Come work with us! If you're interested, apply before Feb 28.

Variational Flow Matching goes Riemannian! 🔮

In this preliminary work, we derive a variational objective for probability flows 🌀 on manifolds with closed-form geodesics, and discuss some interesting results.

Dream team: Floor, Alison & Erik (their @ below) 💥

📜 arxiv.org/abs/2502.12981
🧵1/5

Come to #accelerate25 the premier conference on #aiformaterials #selfdrivinglabs #chemsky Please RT

Very excited about this line of work by Jacco, which opens up opportunities for more scalable modeling of fluids at the mesoscopic scale. I think this will be particularly promising in the context of applications to materials design!

🚨 Excited to share our work just published in Physical Review Letters with @wellingmax.bsky.social, @jwvdm.bsky.social, @berndensing.bsky.social, Marjolein Dijkstra and René van Roij: doi.org/10.1103/Phys....

Details below 👇

I think that a key difference here may be that RPG mechanics are generally less stochastic than the conference reviewing process.

Join us on 27 Feb in Amsterdam for the ELLIS workshop on Representation Learning and Generative Models for Structured Data ✨

sites.google.com/view/rl-and-...

Inspiring talks by @eisenjulian.bsky.social, @neuralnoise.com, Frank Hutter, Vaishali Pal, TBC.

We welcome extended abstracts until 31 Jan!

Postdoc In Organic/Medicinal Chemistry for generative AI in Drug Discovery De TU/e is voortdurend op zoek naar wetenschappelijk en niet-wetenschappelijk personeel om haar ambities waar te maken. Kijk hier voor ons actuele vacatureaanbod.

Open #postdoc position in *Organic/Medicinal Chemistry* for generative #AI in Drug Discovery — join us @tueindhoven.bsky.social 💪🏻

Funded by the European Research Council (ERC).
Deadline: Jan 15, 25

jobs.tue.nl/nl/vacature/...

Repost appreciated!

Really enjoyed this talk from Arnaud! It is so refreshing when a speaker focuses on providing a lucid overview of methods and their connections, past and present. Something to aspire to for all of us.

It's located near the west entrance to the west side of the conference center, on the first floor, in case that helps!

When a bunch of diffusers sit down and talk shop, their flow cannot be matched😎

It's time for the #NeurIPS2024 diffusion circle!

🕒Join us at 3PM on Friday December 13. We'll meet near this thing, and venture out from there and find a good spot to sit. Tell your friends!

If you're at #NeurIPS2024, join us this afternoon to talk diffusion, flows and all that jazz. Be there or be non-circular!

Here are the dapper @metodjazbec.bsky.social and Alexander Timans, presenting their work on early-exit with risk control.

François Cornet and @gbarto are presenting their work on Equivariant Neural Diffusion now at East Exhibition #2403!

Come talk to @zmheiko.bsky.social about his work on sample-efficient black box variational inference! (East Exhibition Hall #4105)

VISA: Variational Inference with Sequential Sample-Average Approximations

Heiko Zimmermann · Christian Andersson Naesseth · Jan-Willem van de Meent

Thu 12 Dec 11 a.m. PST — 2 p.m. PST
East Exhibit Hall A-C #4105

Today my collaborators are presenting no less than 3 of our papers at #NeurIPS2024 Come check out the posters and talk to everyone to find out more about molecule generation, early-exit with risk control, or speeding up variational inference!

Best quote from @eijkelboomfloor.bsky.social so far: "it is aggressively simple!"