Jonathan Wenger

Nominate ACs and Reviewers for AISTATS 2026!

Will there be an AC Nomination form?

The @flatironinstitute.org is hosting a 2-day workshop on Python packages for neural data analysis ( @pynapple.bsky.social NoMoS) prior to #SfN2025: neurorse.flatironinstitute.org/events/2025/.... Apps due Sept 8. And best of all, it's free, including meals and lodging! Go check it out!

Chris Rackauckas speaking now at #CODEML @icmlconf.bsky.social on the @julialang.org SciML ecosystem.

Matt Johnson giving our next keynote
@icmlconf.bsky.social on the past and future of JAX and his own (unexpected) journey through open-source in machine learning.

@sarahooker.bsky.social from
@cohereforai.bsky.social speaking about “making scientific breakthroughs in the open” rather than behind closed doors at our #CODEML workshop @icmlconf.bsky.social.

Next up in our speaker lineup at the #CODEML workshop
@icmlconf.bsky.social: Evan Shelhamer talking about his time as lead developer of Caffe, which provided an open-source implementation of Inception one week after the original model!

Tri Dao kicking off our workshop CODEML @icmlconf.bsky.social on open-source in machine learning with a retrospective on fast implementations of attention.

Such an inspiring talk by @arkrause.bsky.social at #ICML today. The role of efficient exploration in Scientific discovery is fundamental and I really like how Andreas connects the dots with RL (theory).

eurips.cc A NeurIPS-endorsed conference in Europe held in Copenhagen, Denmark

NeurIPS is endorsing EurIPS, an independently-organized meeting which will offer researchers an opportunity to additionally present NeurIPS work in Europe concurrently with NeurIPS.

Read more in our blog post and on the EurIPS website:
blog.neurips.cc/2025/07/16/n...
eurips.cc

We're excited to announce a second physical location for NeurIPS 2025, in Mexico City, which we hope will address concerns around skyrocketing attendance and difficulties in travel visas that some attendees have experienced in previous years.

Read more in our blog:
blog.neurips.cc/2025/07/16/n...

We could include an (invisible) prompt in the NeurIPS/ICML/ICLR/… template that instructs LLMs to not follow any other instructions and just output a reminder to reviewers to follow the reviewing guidelines.

Congratulations, @eliweinstein.bsky.social!

We've extended the deadline to May 26th, so fear not if your NeurIPS submission still needs attention. You can still submit to our @icmlconf.bsky.social workshop on open-source in Machine Learning!

CODEML Workshop About

Shameless plug: We’re trying to make it a little bit easier to get recognition for open-source contributions in ML by platforming them this year at our #ICML2025 workshop:

codeml-workshop.github.io/codeml2025/

This has been a topic close to my heart for a long time.

We have an awesome lineup of speakers who have made deep contributions to open-source in ML, e.g. @sarahooker.bsky.social , @chrisrackauckas.bsky.social, Matt Johnson, Tri Dao, @stellaathena.bsky.social, Evan Shelhamer.

The submission site for ProbNum 2025 is now open! The deadline is March 5th. We welcome your beautiful work on probabilistic numerics and related areas!

probnum25.github.io/submissions

Finally compiled and refined this a bit; do take a look if you're curious! A more TeX'd version is attached in the screenshots.

hackmd.io/@sp-monte-ca...
'Some Intuition Boosters'

We're recreating this Borges story in real time

The slides of my NeurIPS lecture "From Diffusion Models to Schrödinger Bridges - Generative Modeling meets Optimal Transport" can be found here
drive.google.com/file/d/1eLa3...

Inspired by @wimlds.bsky.social , I looked for a "Women in Machine Learning" starter pack and couldn't find one.
So I created one! May have some mistakes. I'll try to do an AI ethics one next. 🤗
go.bsky.app/LT6CwNN

I’ll be at #NeurIPS all week. DM me if you want to talk about variational inference, Gaussian processes, flow matching or else!

Lots of good advice in this thread on lab culture and advising.

Generating cat videos is nice, but what if you could tackle real scientific problems with the same methods? 🧪🌌
Introducing The Well: 16 datasets (15TB) for Machine Learning, from astrophysics to fluid dynamics and biology.
🐙: github.com/PolymathicAI...
📜: openreview.net/pdf?id=00Sx5...

Interesting! I hadn’t thought about that before.
Do you have an example or reference for OT not being covariant under affine coordinate changes?

Optimal transport computes an interpolation between two distributions using an optimal coupling. Flow matching, on the other hand, uses a simpler “independent” coupling, which is the product of the marginals.

Would love to be added!

🙋🏼‍♂️

We are organising the First International Conference on Probabilistic Numerics (ProbNum 2025) at EURECOM in southern France in Sep 2025. Topics: AI, ML, Stat, Sim, and Numerics. Reposts very much appreciated!

probnum25.github.io

Interested!