🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Great opportunity to do a postdoc with us!

Congratulations, Gianluigi Lopardo. Very nice work! It was a pleasure to be on your committee.

Great talk today by Robi Bhattacharjee Robi Bhattacharjee in the Theory of Interpretable AI seminar about how "Auditing Local Explanations is Hard"! Recording available at youtu.be/40XZ_QAU1Mw Upcoming seminar talks: tverven.github.io/tiai-seminar/ Ulrike Luxburg Michal Moshkovitz Suraj Srinivas

Really enjoyed speaking with Greg on unSILOed Podcast with Greg LaBlanc about the Atomic Human.

Really appreciate this writeup of my and Sayash Kapoor's views on AI safety by Tim B. Lee (Timothy B. Lee). understandingai.org/p/six-principl… Tim has been making many similar points on his excellent Substack. We are currently wrapping up a paper that elaborates on these views, so we'll have a

Cool stuff

For PhD students in AI & Mathematics in NL: AIM is organizing a networking event again on Nov. 21 at CWI . Highly recommended opportunity to present your work and meet other PhDs! Sign up here: aimath.nl/index.php/2024…

PLS SHARE: I'm hiring a PhD student to work on ML theory, to begin in Fall 2025. Topics include: generalization bounds & statistical inference via online prediction, representation learning via optimal transport, sequential decision making... More info: cs.bme.hu/~gergo/jobs.ht…

COLT 2025 Call for Paper is out! learningtheory.org/colt2025/ Submit your awesome ML theory works to COLT 2025 (Feb 6 deadline) and save the date for the conference in Lyon, France (June 30-July 4)! #COLT2025

Trying out b*sky. Anyone there yet? Find me at tverven.bsky.social (I think).

Looking forward to speaking in this very timely seminar next week ! Thanks to Tim van Erven, Michal Moshkovitz and Suraj Srinivas for inviting !

Andrej's talk is coming up this Monday Nov 18! See tverven.github.io/tiai-seminar/ for clickable zoom link and more seminar info.

Everyone is moving, people!

Interested in transformers and how they work? Andrej Risteski will talk *today* about two simple linguistic tasks to explore exactly that! ⏲️ Monday Nov 18, 4pm CET/10am EST (Notice the unusual day and hour!) Tim van Erven Suraj Srinivas

Yesterday Andrej Risteski gave an excellent talk at the Theory of Interpretable AI Seminar on the interpretability of attention layers. Interestingly, attentions are not always interpretable! If you missed it, here is the recording youtube.com/@TheoryofInter… Tim van Erven Suraj Srinivas

📢Theory of Interpretable AI Seminar📢 On Thursday, Depen Morwani from Harvard will present work on how **margin maximization** can explain observed phenomena in mech. interpretability! ⏲️Thursday Dec 5th, 4pm CET / 10am EST 🌐tverven.github.io/tiai-seminar/ Michal Moshkovitz Tim van Erven

The theory of interpretable AI seminar is back after the holiday season! 🎅🤶 Our next talk is next Thursday by Yishay Mansour who will talk about interpretable approximations 💻 Website: tverven.github.io/tiai-seminar/ ⏰Date: 16 Jan Suraj Srinivas Tim van Erven Yishay Mansour

In case you missed it: here is the recording of Yishay Mansour's talk about the ability of decision trees to approximate concepts: youtu.be/uOwuho2er58 For upcoming talks, check out the seminar website: tverven.github.io/tiai-seminar/