Conference Goal
In recent years, the field of Machine Learning has made significant progress in theory and applications. This success is rooted in the mutual stimulation of mathematical insight and experimental studies. On the one hand, mathematics allows to conceptualize and formalize core problems within learning theory, leading, for instance, to performance bounds for learning algorithms. On the other hand, experimental studies confirm theoretical predictions and instigate new directions in theoretical research. This meeting aims to discuss the interaction between theory and practice, with focus on the current gaps between the two. The talks will be centered around themes including the following.
- Gradient Methods (gradient optimization, stochastic gradient, natural gradients, gradient applied to deep networks, ...)
- Natural Geometric Structures (Information Geometry, optimal transport geometry, ...)
- Generalisation Theory (statistical learning theory, complexity measures, Ill-posed inverse problems, regularization, implicit bias...)
- Functional analytical tools (approximation theory, harmonic analysis, ...)
- Overparametrization and random matrix theory (neural tangent kernel, lazy training, convergence of gradient descent, generalization bounds, ...)
Confirmed Keynote Speakers
- Francis Bach INRIA Paris Centre, France
- Misha Belkin University of California San Diego, USA
- Lénaïc Chizat EPFL, Switzerland
- Kenji Fukumizu Institute of Statistical Mathematics, Japan
- Stefanie Jegelka MIT, USA, and TU Munich, Germany
- Markos Katsoulakis University of Massachusetts Amherst, USA
- Gitta Kutyniok LMU Munich, Germany
- Jürgen Jost * MPI for Mathematics in the Sciences, Germany
- Frank Nielsen Sony Computer Science Laboratories Inc., Japan
- Gabriele Steidl TU Berlin, Germany
- Matus Telgarsky New York University, USA
* Jürgen Jost will deliver a speech dedicated to our keynote speaker Sayan Mukherjee who sadly passed away on March 31, 2025, in Leipzig. He was Humboldt Professor and played a key role in AI research of the Leipzig University and the Max Planck Institute for Mathematics in the Sciences. His death is a big loss. More information by the MPI MiS.
Organisers
- Nihat Ay Hamburg University of Technology, Germany, and Santa Fe Institute, USA
- Martin Burger DESY and University of Hamburg, Germany
- Benjamin Gess TU Berlin and MPI for Mathematics in the Sciences, Germany
- Guido Montúfar UCLA, USA, and MPI for Mathematics in the Sciences, Germany
Scientific Committee
- Alexandra Carpentier University of Potsdam, Germany
- Asja Fischer Ruhr University Bochum, Germany
- Armin Iske University of Hamburg, Germany
- Stefanie Jegelka MIT, USA, and TU Munich, Germany
- Parvaneh Joharinad Leipzig University and MPI for Mathematics in the Sciences, Germany
- Sören Laue University of Hamburg, Germany
- Johannes Lederer University of Hamburg, Germany
- Marco Mondelli Institute of Science and Technology, Austria
- Sayan Mukherjee * Leipzig University, Germany
- Liza (Elizaveta) Rebrova Princeton University, USA
- Gabriele Steidl TU Berlin, Germany
- Diaaeldin Taha MPI for Mathematics in the Sciences, Germany
- Melanie Weber Harvard University, USA
- Jens-Peter M. Zemke Hamburg University of Technology, Germany
Conference Program
For the detailed list of conference program, talks and posters, please visit here.
Application for Participation and Presentation
If you would like to participate in the conference, you are kindly invited to apply here, where you can also submit a title and an abstract (up to 2 pages, pdf) in case you would like to contribute a talk or/and a poster. The application deadline is May 4th, 2025. A notification email about the outcome of your application will be sent out in early June, 2025. The registration fee for accepted participants is 60 Euros.
We have a limited number of travel grants for student participants. If you are interested in applying for travel support please indicate this in your application statement and provide contact information of a reference person.
For registration and submission issues, contact Hojat Naghavi: dsf-it(at)tuhh.de
