Lightnews — Scholar-powered news

psteinb.bsky.social @psteinb.bsky.social · Aug 19

Looky Looky! 😍🥳👏
arxiv.org/abs/2508.12939
Super fun project, I ❤️ed coauthoring w/ @sbi-devs.bsky.social.
Great lead by @deismic.bsky.social & @janboelts.bsky.social. Contribs by many talented people @jakhmack.bsky.social. 🙏 to #BenjaminKurtMiller for the kickstart! @helmholtzai.bsky.social

Simulation-Based Inference: A Practical Guide

A central challenge in many areas of science and engineering is to identify model parameters that are consistent with prior knowledge and empirical data. Bayesian inference offers a principled framewo...

arxiv.org

9 39

psteinb.bsky.social @psteinb.bsky.social · Jun 3

The 0th day of #HAICON25 was a blast! I've had the plsre to participate in a tutorial on @sbi-devs.bsky.social by #AlinaBazarova and #JoseRoblero. @helmholtzai.bsky.social @fz-juelich.de In the afternoon, I ran a birds-of-a-feather on inverse problems in Helmholtz. @hzdr.bsky.social

1 4

Reposted

RAV @rav-ev.bsky.social · Jun 1

Heute vor 6 Jahren wurde Walter #Lübcke von einem Rechtsradikalen ermordet. Walter Lübcke steht für einen Konservatismus, der heute so sehr fehlt. Ein Konservatismus, der humane Werte, das Menschenrecht auf Asyl, offensiv verteidigt. Und der nicht opportunistisch Positionen der #NoAfD übernimmt.

15 290 1.3K

psteinb.bsky.social @psteinb.bsky.social · May 19

From anecdotal evidence: same holds for the German schools, I have contact with.

Scott McGrath @smcgrath.phd · May 19

Public records from early 2023 reveal US schools were largely unprepared for ChatGPT's rapid adoption by students. Responses varied widely, highlighting the lack of a unified strategy as the technology quickly became a significant challenge in education.
#MLSky #AcademicSky #EduSky

American Schools Were Deeply Unprepared for ChatGPT, Public Records Show

Thousands of pages of documents show school districts around the country did not understand how much ChatGPT would change their classrooms, and pro-AI consultants filled in some of the gaps.

www.404media.co

psteinb.bsky.social @psteinb.bsky.social · Apr 29

Cunning idea by the looks of it!

Robert Haase @haesleinhuepf.bsky.social · Apr 25

Are you a #BioImageAnalysis pro? Challenge your skills by playing #BIAbubbles! 🎉
Find the right workflows and reproduce the result! 🔬🖥️🕹️
And if you have challenging BIA tasks, new level contributions are welcome! 😀
github.com/haesleinhuep...

1

psteinb.bsky.social @psteinb.bsky.social · Apr 26

@learnbayesstats.bsky.social if you like to explore the field of simulation-based inference more, people like @kylecranmer.bsky.social @glouppe.bsky.social @jakhmack.bsky.social or people of the @sbi-devs.bsky.social project could be really good next guests. Lots of things happening there.

5

psteinb.bsky.social @psteinb.bsky.social · Apr 26

I'm done listening. As always, @vincefort.bsky.social is a super knowledgeable guest and a good communicator. Some parts (PAC-Bayes, amortized inference) are sometimes hard to follow due to high density of math/terms. Still very enjoyable and fun episode.

2 2

psteinb.bsky.social @psteinb.bsky.social · Apr 11

🥳Great news, our JOSS paper "sbi reloaded" has been accepted! 🎉
This community lead by the fine folks of @sbi-devs.bsky.social is very welcoming and super fun to work with! I learn with every discussion I have.
paper: joss.theoj.org/papers/10.21...
review: github.com/openjournals...

[REVIEW]: sbi reloaded: a toolkit for simulation-based inference workflows · Issue #7754 · openjournals/joss-reviews

Submitting author: @janfb (Jan Boelts) Repository: https://github.com/sbi-dev/sbi Branch with paper.md (empty if default branch): joss-submission-2024 Version: v0.24.0 Editor: @boisgera Reviewers: ...

github.com

3 6

psteinb.bsky.social @psteinb.bsky.social · Apr 11

Interesting paper! Thanks for sharing. And from what it looks like, no prompt engineering was done. 🤷 This likely leaves the LLM performance under the table. 🐌 To me, this adds apples-oranges (🍎vs🍊) vibe to the robust & well-done stats analysis. Plus, only #openai models were tested. 🔍

2

psteinb.bsky.social @psteinb.bsky.social · Apr 3

Can’t wait to listen in!

Pierre-Simon Laplace @learnbayesstats.bsky.social · Apr 3

What if AI could know when it doesn’t know?

🎙️ @alex-andorra.bsky.social talks with @vincefort.bsky.social about Bayesian deep learning, why it matters for uncertainty, calibration & more

🎧 Tune in: learnbayesstats.com/episode/129-...

#BayesianDeepLearning #MachineLearning #ReliableAI #AIResearch

1 2

Reposted

Riccardo Sapienza @riccard0.bsky.social · Apr 3

I also agree that openness is they key skill for a researcher

www.nature.com/articles/s41...

Openness guides discovery - Nature Biotechnology

Nature Biotechnology - Openness guides discovery

www.nature.com

2 3

Reposted

Serge Belongie @serge.belongie.com · Mar 30

Would you present your next NeurIPS paper in Europe instead of traveling to San Diego (US) if this was an option? Søren Hauberg (DTU) and I would love to hear the answer through this poll: (1/6)

NeurIPS participation in Europe

We seek to understand if there is interest in being able to attend NeurIPS in Europe, i.e. without travelling to San Diego, US. In the following, assume that it is possible to present accepted papers ...

docs.google.com

6 160 280

psteinb.bsky.social @psteinb.bsky.social · Mar 31

Great initiative to see if Neurips could be replicated in Europe! 🇪🇺

Serge Belongie @serge.belongie.com · Mar 30

Would you present your next NeurIPS paper in Europe instead of traveling to San Diego (US) if this was an option? Søren Hauberg (DTU) and I would love to hear the answer through this poll: (1/6)

NeurIPS participation in Europe

We seek to understand if there is interest in being able to attend NeurIPS in Europe, i.e. without travelling to San Diego, US. In the following, assume that it is possible to present accepted papers ...

docs.google.com

psteinb.bsky.social @psteinb.bsky.social · Mar 31

Great to see that many noise2void inspired denoising algorithms have been ported to @pytorch.org! Kudos to @florianjug.bsky.social and many others ...
github.com/CAREamics/ca...

GitHub - CAREamics/careamics: A deep-learning library for denoising images using Noise2Void and friends (CARE, PN2V, HDN etc.), with a focus on user-experience and documentation.)

A deep-learning library for denoising images using Noise2Void and friends (CARE, PN2V, HDN etc.), with a focus on user-experience and documentation.) - CAREamics/careamics

github.com

9 18

psteinb.bsky.social @psteinb.bsky.social · Mar 21

It's been a blast, thanks to @sbi-devs.bsky.social ! This week's hackathon was phenomenal! 🙏 😍 The sbi hackathon welcomed about 25 people in Tübingen with contributions spanning the globe , e.g. 🇺🇸🇯🇵🇧🇪🇩🇪. Wanna see, what we did? Check out the PRs👇
github.com/sbi-dev/sbi/...

Pull requests · sbi-dev/sbi

sbi is a Python package for simulation-based inference, designed to meet the needs of both researchers and practitioners. Whether you need fine-grained control or an easy-to-use interface, sbi has ...

github.com

6 14

psteinb.bsky.social @psteinb.bsky.social · Mar 17

This week will be an upskill from dusk till dawn:
I'll be at the hackathon of @sbi-devs.bsky.social ! I am so much looking forward to meet and work with collaborators. Learn a lot and push #sbi forward. Can't wait to arrive🚆! @helmholtzai.bsky.social @tuebingen-ai.bsky.social #appliedAI

1

psteinb.bsky.social @psteinb.bsky.social · Mar 15

What a gem! Lovely article on our blind focus on accuracy in ML. Can't wait to get my hands on the references therein.

Vincent Fortuin @vincefort.bsky.social · Mar 13

I wrote a little blog post for the Neptune.ai blog about our ICML position paper on Bayesian Deep Learning: neptune.ai/blog/bayesia...

Bayesian Deep Learning is Needed in the Age of Large-Scale AI [Paper Reflection]

In particular, we point out that there are many other metrics beyond accuracy, such as uncertainty calibration, which we have to take into account to ensure that better models also translate to better...

neptune.ai

1 1 3

psteinb.bsky.social @psteinb.bsky.social · Mar 14

Working with #julia sometimes feels like playing minesweeper. Found a helpful library on the web 💣, punched in `add AwesomeLib` and boom 💥 I have to wait for 15min 🥱 to get all dependencies compiled and installed. #21century #oldhardware

psteinb.bsky.social @psteinb.bsky.social · Mar 12

Congrats! 👏

ayakimovich.bsky.social @ayakimovich.bsky.social · Feb 28

Our new press release is out! www.hzdr.de/db/Cms?pOid=... @hzdr.bsky.social @hzdr.bsky.social @helmholtzai.bsky.social 🧪

From handicap to asset: AI approach leverages optics phenomenon to produce better images - Helmholtz-Zentrum Dresden-Rossendorf, HZDR

Quantitative phase imaging (QPI) is a microscopy technique widely used to investigate cells and tissues. Even though first biomedical applications based on QPI have been developed, both acquisition sp...

www.hzdr.de

1

psteinb.bsky.social @psteinb.bsky.social · Mar 12

I think, this is a great opportunity for anyone interested!

Mathieu Seyfrid @mathieuseyfrid.bsky.social · Mar 11

Thrilled about the launch of our first Helmholtz AI 🇩🇪x Mila 🇨🇦 call, for PhD students of Helmholtz AI Associates to visit groups at @mila-quebec.bsky.social and pollinate new connections with the @www.helmholtz.de Association: cross-field, cross-center. @helmholtzai.bsky.social

psteinb.bsky.social @psteinb.bsky.social · Mar 10

🗓️ No plans for Wednesday in Dresden, March 12, 5:30pm yet?

Then please join us for a fun event on Computer Vision 💻 and Image analysis 🔬 🔭 ... there will be food for thought 🤯 , drinks 🥛 and pizza 🍕 .

For the agenda and registration see the link below 👇

events.hifis.net/event/2248/

events.hifis.net

2

psteinb.bsky.social @psteinb.bsky.social · Mar 6

Jaw dropping results! Congrats @maxxxzdn.bsky.social for this wonderful treat. Happy to put this into play.

Max Zhdanov @maxxxzdn.bsky.social · Mar 5

🤹 Excited to share Erwin: A Tree-based Hierarchical Transformer for Large-scale Physical Systems

joint work with @wellingmax.bsky.social and @jwvdm.bsky.social

preprint: arxiv.org/abs/2502.17019
code: github.com/maxxxzdn/erwin

1

psteinb.bsky.social @psteinb.bsky.social · Feb 21

The @turingpost.bsky.social send out a survey to its readers. Starting the thing, I already felt excluded a bit 👇👇👇
But hey, being at #Neurips24 already gave be impression that being in academia and working on #AI is a niche profession these days. 🤷

1

psteinb.bsky.social @psteinb.bsky.social · Feb 17

Is it time to buy a new laptop, if #julia crashes with an OOM error during project build if it can only use 6GB of RAM? #PrematureOptimisation #InvertedReality

psteinb.bsky.social @psteinb.bsky.social · Feb 14

🎉We started with some colleagues of @esciencecenter.bsky.social and 🏴󠁧󠁢󠁷󠁬󠁳󠁿 to devise a @carpentries.carpentries.org lesson on #DeepLearning. Based on the incredible commitment of many individuals, the lesson is now adopted into the carpentries lab! Congrats to all involed! 🥳
github.com/carpentries-...

[Review]: Introduction to deep learning · Issue #25 · carpentries-lab/reviews

Lesson Title Introduction to deep learning Lesson Repository URL https://github.com/carpentries-incubator/deep-learning-intro Lesson Website URL https://carpentries-incubator.github.io/deep-learnin...

github.com

1 3 12