The video recording of Donya's talk is up: youtu.be/r39iRaWymT8

Just learned about this website, with open mathematical problems in machine-learning-adjacent areas such as statistics, learning theory, and optimization. E.g. many open problems from COLT are there. Cool stuff. solveall.org

Join us this Thursday for the Theory of Interpretable AI Seminar! Donya Saless will present her COLT 2025 work on using small subsets of training data to provide robust, compact explanations for ML predictions. 16:00 CET | 10:00 EST tverven.github.io/tiai-seminar/

The Three Laws of Reviewing: 1. A reviewer may not decrease the conference quality or, through inaction, allow the conference quality to come to harm. 2. A reviewer must obey the orders given them by the program chairs except where such orders would conflict with the First Law. 3. ...

The video for Nir Shavit's talk is up: youtu.be/G6BIBqzPoFY

We have Nir Shavit speaking at the Theory of Interpretable AI seminar this Thursday! 🌟 Nir is an MIT professor and Gödel + Dijkstra Prize winner for work in distributed computing. He uses combinatorial ideas to explain neural computation via feature channels. tverven.github.io/tiai-seminar/

List of accepted papers at #COLT2026, the Annual Conference on Learning Theory: learningtheory.org/colt2026/acc... (h/t @gautamkamath.com)

Bingo! Do I win a prize now?

Indeed, a must-read! An existential debate about mathematics is coming up, and now is the time to articulate the many values of mathematics. If mathematicians are seen as machines that turn coffee into theorems, we will be in a lot of trouble.

Summer school on Decisions and Theory for AI! July 13-17 Leiden University Lecturers: Gergely Neu @neu-rips.bsky.social Claire Vernade @claireve.bsky.social Marco Avella Medina Peter Grünwald @petergrunwald.bsky.social Myself Application deadline: April 30 sites.google.com/view/dtailsu...

"AI amplifies whatever is already there. Good discipline becomes great output. No discipline becomes technical debt at machine speed. Anthropic chose a direction. Go faster. Have Claude check Claude. And when it breaks, go faster still." substack.com/home/post/p-...

How are AI tools changing our writing process? In our #chi2026 paper, we find a shift in the way writers write and engage with ideas: from generating ideas to reacting to the ones AI surfaces. We call it reactive writing.

Looking forward to Anton Xue's talk at the Theory of Interpretable AI Seminar! 🚀 He’ll present a framework with stability guarantees for certifying feature attribution explanations and LLM reasoning chains through robust feature selection. Join us! 🧠✨ tverven.github.io/tiai-seminar/

Don't forget this!

We want to speak directly to the concern many of you have expressed, and we owe you a clear explanation of what happened, why it happened, and where we stand now. We understand this situation caused genuine alarm and we take that seriously.

Did I miss the discussion about this over here?

Following the success of the EurIPS and NeurIPS-Mexico City pilots in 2025, we are thrilled to announce two official NeurIPS 2026 satellite events for this year! These will be held in Paris, France and Atlanta, USA, respectively, running alongside the main venue in Sydney, Australia.

Good advice from the university library. Applying this to all my social media as well.

Upcoming talk in the Theory of Interpretable AI seminar: Federico Adolfi frames interpretability in terms of computational complexity theory.

Wow. Donald Knuth on Ai for math: www-cs-faculty.stanford.edu/~knuth/paper...

Very inspiring talk today by Damien in the Theory of Interpretable AI seminar! Recording is up: youtu.be/qI4oVWhvz-4 Upcoming talks on the seminar website: tverven.github.io/tiai-seminar/

Happening today!

Excited for Damien’s seminar talk this Thursday!🚀 Steering is an exciting area in interpretability—but how strongly should we steer? Damien will present a theory of steering strength: choosing the right magnitude of representation change—not too weak, not too strong tverven.github.io/tiai-seminar/

Apparently the NeurIPS chairs decided to reopen the camera-ready submission to ensure "that the proceedings reflect the highest scientific quality.” @neuripsconf.bsky.social, is that true? So the solution to academic misconduct is to give the authors a chance to cover their tracks?

Video recording of Chhavi's talk in the Theory of Interpretable AI seminar is now available: youtu.be/ylW3H--g7SU Check out the seminar website for upcoming speakers: tverven.github.io/tiai-seminar/

Happening today!

Looking forward to the next “Theory of Interpretable AI” seminar on January 15, where Chhavi Yadav will present "ExpProof"! A fresh take on trustworthy explanations for confidential ML models using Zero-Knowledge Proofs. Feel free to join! #interpretability #Crypto tverven.github.io/tiai-seminar/

New blog post (on a shiny new ICML blog!): What's New in #ICML2026 Peer Review Some highlights: - Policies to combat thinly sliced contributions - Cascading desk rejections for peer-review abuse - Reviewer reciprocity - New ways to support authors and reviewers Post: blog.icml.cc/2026/01/08/w...

In case you missed Ayman's talk, here is the video: youtu.be/KunGyRGbRdk

Happening today!

Ayman will present at the Theory of Interpretability Seminar! He’ll share a new approach for optimal, interpretable decision trees with strong efficiency guarantees. 🌳✨ 🗓️ Join us to learn how AO* search leads to provably better sparse trees! tverven.github.io/tiai-seminar/

When reading a large literature, it is really helpful to have opinionated opinions that help to categorize papers. One of mine for explainable AI is that methods need to address a fundamental limit in how much information can be communicated. Blog post: www.timvanerven.nl/blog/xai-com... (no math)

Our libraries are cutting staff so that Elsevier can have its 32% profit margin

The schedule for our Workshop on the Theory of XAI is now online! 🕰️ Dec 2, starting 9am 📍 Bella Center Copenhagen (co-located with EurIPS) 🔗 sites.google.com/view/theory-...

Great seminar talk by @ulrikeluxburg.bsky.social yesterday. Here's the video if you missed it: youtu.be/zR_GvDF65OM Seminar info: tverven.github.io/tiai-seminar/

Happening in two hours.

Coming up tomorrow (Tuesday 11 Nov) in the Theory of Interpretability seminar: Ulrike von Luxburg will discuss why informative explanations only exist for simple functions 👀 tverven.github.io/tiai-seminar/

Here is a formal impossibility result for XAI: Informative Post-Hoc Explanations Only Exist for Simple Functions. I'll give an online presentation about this work next tuesday in @timvanerven.nl 's Theory of Interpretable AI Seminar: arxiv.org/abs/2508.11441 tverven.github.io/tiai-seminar/

The blog post is available: blog.arxiv.org/2025/10/31/a...

I am recruiting via Ellis for a PhD position to develop mathematical foundations for explainable AI. Benefits: - Do new fundamental math / machine learning theory on a societally important topic - Part of a team - Good conditions in Amsterdam, NL Applications via Ellis PhD program before Oct. 31.

The video for Shahaf's talk is now available: youtu.be/AsJB4kym4Tc See the seminar website for upcoming talks: tverven.github.io/tiai-seminar/

Happening tomorrow, October 21!

The next talk in the Theory of Interpretable AI seminar will be by Shahaf Bassan on "The Computational Complexity of Explaining ML Models". Tuesday October 21 tverven.github.io/tiai-seminar/

How can we make AI explanations provably correct — not just convincing? 🤔 Join us for the Theory of Explainable Machine Learning Workshop, part of the ELLIS UnConference Copenhagen 🇩🇰 on Dec 2, co-located with #EurIPS. 🕒 Call for contributions open until Oct 15 (AoE) 🔗 eurips.cc/ellis

🎓 Ready to expand your horizons in #AI research? Do your PhD with 2 leading academic institutions across Europe and build your international network with the Academic Track in the #ELLISPhD Program. Apply now via our central recruiting portal until Oct 31. 👉 All info: https://bit.ly/45DSe75

Looking forward to talking about our work on the value of explanation for decision-making at this workshop

Are you coming ? I will be talking about #XAI #orms

Excellent bachelor's thesis by my student David van Batenburg with a mathematically rigorous treatment of the axiomatic characterization of Shapley values for feature attribution: arxiv.org/abs/2510.03281 Clears up and corrects many math aspects that tripped me up when first learning about SHAP.

Interested in provable guarantees and fundamental limitations of XAI? Join us at the "Theory of Explainable AI" workshop Dec 2 in Copenhagen! @ellis.eu @euripsconf.bsky.social Speakers: @jessicahullman.bsky.social @doloresromerom.bsky.social @tpimentel.bsky.social Call for Contributions: Oct 15

I have 2 open PhD positions on Mathematical Foundations for Explainable AI: Position 1: werkenbij.uva.nl/en/vacancies... (apply by October 13, 2025) Position 2: applications via the Ellis PhD Program: ellis.eu/news/ellis-p... by Oct. 31. Both positions are equivalent (except for starting dates)