New blog post: The Better Lesson? Geometry and Topology in the Era of Deep Learning In which I think a bit about the role fundamental mathematical concepts might yet have to play in the quest for better deep learning architectures. Comments welcome! bastian.rieck.me/blog/2026/be...

Have some (applied) topology research you want to share with the world? Consider adding it to 🍩DONUT🍩, our Database of Original & Non-Theoretical Uses of Topology. Great intro by Barbara Giunti: www.youtube.com/watch?v=eQ_5... 🏠 donut.topology.rocks

For math friends: My work in recent years has been very data science + social justice focused. I have no plans to roll that back, BUT, I have missed doing some capital-M Math, so I've been trying to pump out some mathy papers recently. Here's the latest: Reliable Topology for Dynamic Data. "Enjoy!"

Oh, and to follow up: as a small liberal arts college applied mathematician who writes collaborative papers, it’s common for Math People™ not to take me seriously. I usually avoid solo papers because I find myself boring—but fine. Here you go.

Very proud of my student Juan Amboage, who managed to get his Master thesis accepted at #ICLR2026, with significant help by Ernst Röell and @pseudomanifold.topology.rocks , and very very little input from my side. If you want to know more, here is the ArXiv version: arxiv.org/abs/2510.00757

📢 The submission deadline for ✨GRaM workshop✨ at @iclr-conf.bsky.social is 5th February 2026. 🌟Submit your exciting works in Geometry-grounded representations. We welcome submissions in multiple tracks i.e. 📄 Proceedings 📝extended abstract 👩‍🏫Tutorial/blogpost as well as an exciting challenge!

AI is a new idea? Not quite! In this short blog post, geared towards a general audience, I take a brief tour through history, starting with Homer and ending with ChatGPT. Enjoy! bastian.rieck.me/blog/2025/new/

For more details check out our website: http:// gram-workshop.org or follow @gram-org.bsky.social

GRaM 2.0 is being organised by an ✨incredible team✨ @erikjbekkers.bsky.social @alisometry.bsky.social @sharvaree.bsky.social @oumarkaba.bsky.social @jmtomczak.bsky.social @sukjulian.bsky.social Hannah Lawrence, Robin Walters, Manuel Lecha, Elizabeth Baker and Stefanie Jegelka.

📢The second edition of ✨GRaM workshop✨ is here this time at #ICLR26. 🌟Submit your exciting works in Geometry-grounded representations. We welcome submissions in multiple tracks i.e. 📄 Proceedings 📝extended abstract 👩‍🏫Tutorial/blogpost as well as an exciting challenge!

Geometry & topology may not save the world, but they may well save something that is homotopy-equivalent to it.

Celebrating the author contributions of our ELLIS Members and PhD Students at @neuripsconf.bsky.social 2025 - congratulations to all authors! NeurIPS is joined for the first time by @euripsconf.bsky.social in CPH 🇩🇰, a NeurIPS-endorsed European venue co-developed by ELLIS. 🔗 https://bit.ly/48cRAib

My thesis project embarks on this question, under the supervision of both @erikjbekkers.bsky.social , @pseudomanifold.topology.rocks and @olgatticus.bsky.social we devise a metric over graphs, 𝘵𝘩𝘳𝘰𝘶𝘨𝘩 the maps that exists between them. It also has a cool name! 𝐇𝐨𝐦𝐨𝐦𝐨𝐫𝐩𝐡𝐢𝐬𝐦 𝐃𝐢𝐬𝐭𝐨𝐫𝐭𝐢𝐨𝐧 (ⒽⒹ)

All these insights resulted in a 9.5/10 thesis grade (whatever that means), which I hope serves as a hook to give it a read! 📖 On arxiv: arxiv.org/abs/2511.03068

Under mild assumptions, we obtain a metric and we show 𝘤𝘰𝘮𝘱𝘭𝘦𝘵𝘯𝘦𝘴𝘴. However, as all graphs enthusiasts might notice, this implies the GI problem. That's not the end of the story because we show that we can sample graphs and achieve completeness in expectation :0 !

Two month flew by since the start of the PhD journey with @pseudomanifold.topology.rocks and now the time is right to share how our collaborations officially started, with a question: How can we measure difference in (attributed) graphs beyond binary isomorphism testing ? 🌀

Beware of low-hanging fruits in research, lest you eat the forbidden fruit and get expelled from paradise. #showerthoughts #research #academia

Excited to be at #SfN25! Come catch my poster “Neural dimensionality expands over the course of brain-computer interface learning” wed PM (W15) if you’re interested in non-invasive BCIs, neural manifolds, human learning, etc. Building upon our recent work (doi.org/10.1101/2025.03.29.646109)

Big graph, little compute? In our #NeurIPS2025 paper we propose geometry-aware edge-contraction-based pooling methods for GNNs. Our methods preserve graph structure, make interpretable pooling choices, and ensure robust performance at downstream tasks. 🧵1/n

Now with an excellent YouTube video by @ben300694.bsky.social: youtu.be/Vkl6dLKnnJc

New blog post: Skills for the Future A brief reflection on what skills may (hopefully) remain relevant in the new (hopefully bright) future of LLMs et al. #AcademicWriting #AcademicChatter bastian.rieck.me/blog/2025/sk...

Interesting application of ideas from topology to LLM training!

EurIPS student pricing is ending soon and I have a hunch that the other tickets might also sell out, so get yours now while supplies last!

It's epoch 50. Do you know whether your LLM is grokking? In our #NeurIPS2025 paper "Less is More: Local Intrinsic Dimensions of Contextual Language Models," we study how the local intrinsic dimension (LID) of embeddings changes during training and fine-tuning. #research 🧵1/n

Want to generate point clouds? Our (w/ Ernst Röell) #NeurIPS2025 paper shows a topology-driven method that is fast, simple, and high-quality. 🔥 1/n

Oh, you have criticisms of my paper? Surely you noticed the part of the paper where we already rebutted your critique, by writing “Possible criticisms: the methodology of this paper is bad. We are aware of this.”

Currently, on Hacker News.

Will AI Replace Human Thinking? The Case for Writing and Coding Manually.

(if you're really curious, we have a full github repo for the paper here: github.com/dkobak/low-d...)

one of the more fun topics stemmed from a discussion with @hippopedoid.bsky.social over what the oldest 2D PCA visualization we could find was. After some scouring, we settled on a 1960 paper from researchers at the Université de Montréal about turtle carapaces, which we recreated.

Last year I met a bunch of great researchers who work with high-dimensional data at a Dagstuhl seminar. This week we put out a preprint about the history and philosophy of low-dimensional embedding methods, their applications, their challenges, and their possible future arxiv.org/abs/2508.15929

Hope you all had a good summer. I'm very happy to announce the speaker line-up for the falls Chalmers AI4Science seminars! Hope to catch you all there!

New blog post: Consciousness Is Overrated Anyway (for AI) In which I, somewhat coherently, try to contribute to the whole "Is AI Conscious" debate. 🔗 bastian.rieck.me/blog/2025/co... #AI #MachineLearning

Should we still teach machine learning classes? Let me justify my existence.

Wonder when I'll reach the stage of my career where I just sign with an initial... People who do that already, how cool do I have to be to join the club? 🥺 #AcademicSky #MathSky

Come find me (and this lovely bin of recyclable pens) later today at 181! @cogcompneuro.bsky.social

Save the Date! ATMCS 12 is coming up fast, January 26-30, 2026, in Leipzig. More info at the link below, call for contributions coming soon! www.mis.mpg.de/events/serie...

📢 Save the date! Join us for the next @ellis.eu x UniReps Speaker Series! 📅 27th August – 16:00 CEST 📍https://ethz.zoom.us/j/66426188160 🎙️ Speakers: Keynote by @lelandmcinnes.bsky.social & Flash Talk by Yu (Demi) Qin 🔔 Stay updated by joining our Google group: groups.google.com/u/2/g/ellis-...

🎥 Shapes, Spaces, Simplices, and Structure: Geometry, Topology & Machine Learning "What if the answer to some problems in graph learning is not more, but better structure?" This is the central premise of my talk @logml.bsky.social and @unireps.bsky.social. 🧵1/5

"SAFE TY FIRST" Kerning. Once you know about it, your enjoyment of public signage will be ruined forever. en.wikipedia.org/wiki/Kerning

Happy to announce I just defended my dissertation at Harvard and I’m now a PhD.

🇨🇭 Happy #SwissNationalDay! It's already a year since I started my new position @unifr.bsky.social and I'v had the good fortune to be living here since 2018. Thanks for everything, Switzerland! 🙏 Here's to the future! 🚀

While you are waiting for your #NeurIPS reviewers, you might enjoy this tongue-in-cheek blog post I wrote: The Missing Screwtape Letters from Academia bastian.rieck.me/blog/2023/sc... (With apologies to C.S. Lewis)

We grow better and stronger: a few inspiring visits between Mila and Helmholtz Munich: 1. read below on what a few week stays at Mila brought to Irene and Gab, PhD students from Benjamin Schubert and Maria Colome Tatché 's groups @HelmholtzMunich […] [Original post on wisskomm.social]

Looking very much forward to this event! Thanks for having me!

🔥 Mark your calendars for the next session of the @ellis.eu x UniReps Speaker Series! 🗓️ When: 31th July – 16:00 CEST 📍 Where: ethz.zoom.us/j/66426188160 🎙️ Speakers: Keynote by @pseudomanifold.topology.rocks & Flash Talk by @florentinguth.bsky.social

Topology is cool, check it out!

#MachineLearning loves data. #Topology loves shape. Let's bring them together! 👉The Euler Characteristic Transform (ECT) bridges both worlds, as I outline in a recent article in the Notices of the @amermathsoc.bsky.social Engage! 🧵1/n

Join us on a (mathematical) journey to a shire - oops, HIGHER - standard and principled evaluation schema for our benchmark datasets. This is the reward of the RINGS framework. 📒 Blog: aidos.group/blog/rings/ 📃 Paper: doi.org/10.48550/arX... 👩‍💻 Code: github.com/aidos-lab/ri...