Informal learning seminar on mathematics and AI

by Maximilian Schimpf (MPIM)

Friday Jul 3, 2026, 4:00 PM → 5:00 PM Europe/Berlin

0.008 (Mathezentrum)

Title: “Why Transformers can (in theory) do anything”

Let us imagine that there existed an optimal "next word prediction" function \Gamma. For example, if we entered "Is the Riemann hypothesis true?", then \Gamma would output the answer. Assuming that \Gamma exists, we can ask whether it is possible to approximate it with the transformers algorithm. This talk will present a positive answer to this question given in the paper "Transformers are Universal In-context Learners" by Furuya de Hoop and Peyré.

https://sites.google.com/view/informal-math-ai-bonn/home

---