From singularities to statistics: the geometry of learning
- Date
- Oct 22, 2026
- Time
- 3:00 PM - 4:00 PM
- Speaker
- Jiayi Li
- Affiliation
- MPI-CBG/CSBD
- Language
- en
- Main Topic
- Biologie
- Host
- CSBD Math Seminar
- Description
- Resolution of singularities, zeta functions and log canonical thresholds are not the usual tools of statistics. Singular learning theory shows that they are exactly what is needed to describe Bayesian learning in models whose parameters are not identifiable — which includes almost every model used in practice. I will give an account of this theory for an audience of mathematicians: the free energy expansion, the role of Hironaka’s theorem, the real log canonical threshold as a learning coefficient, and the questions that remain open about computing it and extending it beyond the Bayesian setting. Throughout, I will point to the lectures of the SLT Days workshop at MPI-CBG on 26–27 October, which take up each of these threads.
Last modified: Oct 22, 2026, 7:36:26 AM
Location
Max Planck Institute of Molecular Cell Biology and Genetics (MPI-CBG CSBD SR Ground Floor (VC))Pfotenhauerstraße10801307Dresden
- Phone
- +49 351 210-0
- Fax
- +49 351 210-2000
- MPI-CBG
- Homepage
- http://www.mpi-cbg.de
Organizer
Max Planck Institute of Molecular Cell Biology and GeneticsPfotenhauerstraße10801307Dresden
- Phone
- +49 351 210-0
- Fax
- +49 351 210-2000
- MPI-CBG
- Homepage
- http://www.mpi-cbg.de
Scan this code with your smartphone and get directly this event in your calendar. Increase the image size by clicking on the QR-Code if you have problems to scan it.
Legend
- Biology
- Chemistry
- Civil Eng., Architecture
- Computer Science
- Economics
- Electrical and Computer Eng.
- Environmental Sciences
- for Pupils
- Law
- Linguistics, Literature and Culture
- Materials
- Mathematics
- Mechanical Engineering
- Medicine
- Physics
- Psychology
- Society, Philosophy, Education
- Spin-off/Transfer
- Traffic
- Training
- Welcome