🚨 Workshop on the Theory of Explainable Machine Learning Call for ≤2 page extended abstract submissions by October 15 now open! 📍 Ellis UnConference in Copenhagen 📅 Dec. 2 🔗 More info: sites.google.com/view/theory-... @gunnark.bsky.social @ulrikeluxburg.bsky.social @emmanuelesposito.bsky.social

Late, but arxiv.org/abs/0804.2996 is *incredible*, so many good lines (e.g., "This comes close to being an accusation of a false claim of priority for a false discovery of an untrue fact, which would be a rare triple-negative in the history of intellectual property disputes.").

I am hiring PhD students and/or Postdocs, to work on the theory of explainable machine learning. Please apply through Ellis or IMPRS, deadlines end october/mid november. In particular: Women, where are you? Our community needs you!!! imprs.is.mpg.de/application ellis.eu/news/ellis-p...

I reported a paper with fake citations to the Neurips editors a few months ago, and I've been informed Neurips will be introducing systematic checking of citations.

Many interesting ideas in Tobias' talk today! You can still watch the recording: youtu.be/IZ4m_II3DVI

Wanna chat with an experienced learning theory researcher? We have open office hours online with four awesome folks, Surbhi Goel (@surbhigoel.bsky.social), Gautam Kamath (@gautamkamath.com), Ayush Sekhari, and Lydia Zakynthinou)! Book a slot and ask them anything! let-all.com/officehours....

Happening today!

🚨 The Theory of Interpretable AI Seminar is back after the summer! 🎙️ Tobias Leemann: On Identifiability & Representational Power of Post-Hoc Explanations 📅 Sept 9, 4PM CET / 10AM EST 📍 Online: uva-live.zoom.us/j/87120549999 🔗 Website: tverven.github.io/tiai-seminar/

The Netherlands now also has a call out to poach top international (non-EU) researchers whose academic freedom is threatened. Possibility to apply for 1M euro; need to be nominated by a Dutch research institution, which requires a recognizable strong profile. www.nwo.nl/en/calls/tulip-fund

OpenAI is now in the moderation business

Lol, we need the "AI" that is killing the environment and guzzling untold amounts of resources to save the environment. This is 1984 type of marketing, like "ministry of peace" that is actually the war department.

The unsustainability of long distance flights to conferences is a valid concern. Clearly, meeting international experts is important for high-level research, but dynamic regional events to supplement huge monolithic conferences might even make that easier. We need more experiments like EurIPS asap.

EurIPS is coming! 📣 Mark your calendar for Dec. 2-7, 2025 in Copenhagen 📅 EurIPS is a community-organized conference where you can present accepted NeurIPS 2025 papers, endorsed by @neuripsconf.bsky.social and @nordicair.bsky.social and is co-developed by @ellis.eu eurips.cc

The NeurIPS paper checklist corroborates the bureaucratic theory of statistics.

4th AI & Mathematics in NL workshop in Tilburg. Many cool presentations: aimath.nl/index.php/20... And great people:

Great talks make complicated things so clear they seem easy. This was such a talk: youtu.be/nCR8hWgvwTI?...

Happening today: ⏰⏰ Theory of Interpretable AI Seminar ⏰⏰ Chain-of-Thought: Why does explaining to LLMs using CoT prompting work? Join us on June 3, when Bohang Zhang will dive into the mechanisms behind chain-of-thought prompting — and what makes it so effective tverven.github.io/tiai-seminar/

If you're finishing your camera-ready for ACL or ICML and want to cite co-first authors more fairly, I just made a simple fix to do this! Just add $^*$ to the authors' names in your bibtex, and the citations should change :) github.com/tpimentelms/...

Deadlines for PhD and Postdoc vacancies coming up: applications open until Monday June 2!

⏰⏰ Theory of Interpretable AI Seminar ⏰⏰ Chain-of-Thought: Why does explaining to LLMs using CoT prompting work? Join us on June 3, when Bohang Zhang will dive into the mechanisms behind chain-of-thought prompting — and what makes it so effective tverven.github.io/tiai-seminar/

The 4th AIM workshop on mathematics and AI in NL is coming up in two weeks!, with a very cool program: surfdrive.surf.nl/files/index.... It's still possible to register: aimath.nl/index.php/20... Keynotes: @smaglia.bsky.social (UvA), @emiliocarrizosa.bsky.social (Universidad de Sevilla, Spain)

⏰⏰ Theory of Interpretable AI Seminar ⏰⏰ Chain-of-Thought: Why does explaining to LLMs using CoT prompting work? Join us on June 3, when Bohang Zhang will dive into the mechanisms behind chain-of-thought prompting — and what makes it so effective tverven.github.io/tiai-seminar/

Going to the hospital because I broke my wrist smashing the endorse button: www.understandingai.org/p/i-got-fool...

NeurIPS amateur tip: put the checklist at the end of the appendix, so you forget to include it when cutting off the appendix for the main paper submission.

The tutorials, workshops, and community events for #COLT2025 have been announced! Exciting topics, and impressive slate of speakers and events, on June 30! The workshops have calls for contributions (⏰ May 16, 19, and 25): check them out! learningtheory.org/colt2025/ind...

And the video of Gunnar's talk is up on YouTube in case you missed it: youtu.be/7MrMjabTbuM @gunnark.bsky.social

Curious about feature attribution? SHAP & LIME treat features independently—but features interact! Come hear how to "Disentangle Interactions and Dependencies in Feature Attribution" Tuesday May 6 (today!) 4pm CET, 10am ET tverven.github.io/tiai-seminar/

Finally finished the AI Snake OIL book: ssir.org/books/excerp.... Incredibly uplifting to see @randomwalker.bsky.social and @sayash.bsky.social set the example for a higher-quality debate about AI. Perhaps not the final word, but an excellent shot at what can be said now. Highly recommended!

On June 12-13 the 4th workshop on #Mathematics and #AI is taking place at the University of Tilburg. More information & registration: aimath.nl/index.php/20... Keynote speakers: @smaglia.bsky.social (University of Amsterdam), @emiliocarrizosa.bsky.social (Universidad de Sevilla, Spain)

The AI & Mathematics (AIM) initiative is organizing the 4th workshop on Mathematics and AI. For researchers in the Netherlands working on any mathematical aspects of AI. June 12-13, Tilburg Keynotes by @smaglia.bsky.social and Emilio Carrizosa More info and registration: aimath.nl/index.php/20...

Just enjoyed @mircomutti.bsky.social's seminar talk about interpretable meta-learning of contextual bandit types. The recording is available in case you missed it: youtu.be/pNos7AHGMXw

Happening today! Join us if you want to hear about our take on interpretable exploration for multi-armed bandits. If interested but cannot join, here's the arxiv arxiv.org/abs/2504.04505 Joint work with Jeongyeol, Shie, and @aviv-tamar.bsky.social

Tomorrow we have Mirco Mutti in the Theory of Interpretable AI Seminar: tverven.github.io/tiai-seminar/

Now open: vacancies for a PhD or Postdoc position to develop "Mathematical Foundations for Explainable AI" with me. This is a new research direction that I am very excited about, and which will really start to take off over the next few years. Come join my group: www.timvanerven.nl#open-phd-and...

This open letter will be published tomorrow on the AAUP blog. Signatures will be accepted after that, but a high # of signatures in advance will help the organizers gain media attention that could benefit the effort. Share in ways you believe will get results! sites.google.com/view/we-must...

Haven't made up my mind about this, but it's at least a valid discussion point.

⏰⏰Theory of Interpretable AI Seminar ⏰⏰ In two weeks, April 8, Mirco Mutti will talk about "A Classification View on Meta Learning Bandits"

I’ve put together a short list of opportunities for early career academics willing to come to Europe: www.cvernade.com/miscellaneou... This mostly covers France and Germany for now but I’m willing to extend it. I build on @ellis.eu resources and my own knowledge of these systems.

Couldn't make it to Atticus's talk last week on the current state of interpretability? No worries! Fortunately for you, the talk is rescheduled to Wednesday this week😄 Please do join us! 🌐: tverven.github.io/tiai-seminar/

Join us today for a talk by Atticus Geiger on the current state of (mechanistic) interpretability, and the path moving forward!

Hooray, I received a Vici grant from the Dutch science foundation! Heads up for current PhD students in learning theory: I will have two postdoc positions available in Amsterdam on "learning theory for interpretable/explainable AI" in the coming years. www.uva.nl/shared-conte...

⏰⏰Theory of Interpretable AI Seminar ⏰⏰ Tuesday in two weeks we will hear from Atticus Geiger on "The Current State of Interpretability and Ideas for Scaling Up" Website: tverven.github.io/tiai-seminar/ Date: 4 March, 4pm CET/ 10am ET

The rise of gen AI, backed at every level by the far-right, suggests the arrival of an _anti-information_ age. AI offers a self-reinforcing bullshit loop that, much like a Russian disinfo operation, serves to muddy the information space and create doubt about the very nature of reality.

With Jack Mayo we are organizing a symposium on theory of bandit algorithms on March 10 at the University of Amsterdam. Talks by Jack Mayo, Wouter Koolen, Julia Olkhovskaya, Dirk van der Hoeven and, tentatively, Tor Lattimore. www.timvanerven.nl/events/bandi... has details and (free) registration

Reposting to see if we can get some input on what the community is eager to see in the seminar:

Learning theory people, I need your help: in the definition of (agnostic) PAC learning, is the learning alg allowed to depend on epsilon and delta? Literature seems to say 'no' (www.cs.huji.ac.il/~shais/Under..., doi.org/10.1016/0890..., doi.org/10.1145/7635...), but that's silly... 1/2

Our Theory of Interpretable AI seminar (tverven.github.io/tiai-seminar/) will soon celebrate its one-year anniversary! 🥳 As we step into our second year, we’d love to hear from you! What papers would you like to see discussed in our seminar in the future? 📚🔍

Finishing a paper before a conference deadline is like being stuck inside Zeno's paradox: Can we finish a task with an infinite number of steps in finite time?

Great talk by Jeremias Sulam in the interpretable AI seminar today, connecting feature and concept interpretability to hypothesis testing via E-values! Recording available in our YouTube channel: youtu.be/cx7wTtRdhnA Check out the seminar website for upcoming talks: tverven.github.io/tiai-seminar/

Happening today (Tuesday): feature attribution methods through the lens of statistics!