Center for Interdisciplinary Research (ZiF), Bielefeld University
04.08.21 07.08.21

Conference on Mathematics of Machine Learning

The conference will take place in a hybrid format, allowing both in person and online attendance. The submission of contributed talks and posters is now closed.

In recent years, machine learning algorithms have seen an unprecedented success in a variety of fields, ranging from science and engineering to medicine and social sciences. As a result, artificial intelligence has been identified as one of the key technologies for future social and economic advance. With its rising importance also in failure sensitive systems, for example in medical devices and autonomous driving, the need for a systematic understanding of the functionality of machine learning algorithms and for guarantees on their accuracy and precision becomes vital. This calls for the development of a mathematical understanding of the structures and mechanism underlying the success of machine learning techniques. This endeavor has seen a significant increase of interest and efforts in recent years, crucially relying on interdisciplinary interaction, communicating mathematical progress and empirical research. The proposed conference is aimed as a contribution to this rapidly developing field, by bringing together experts from various mathematical areas with shared interest in applications to machine learning and experts from fields such as computer science and biology.

Please see also the poster overview at Poster Session


Yasaman Bahri

Google Brain, USA

Pierre Baldi

University of California, Irvine, USA

Peter Bartlett

University of California, Berkeley, USA

Minh Hà Quang


Matthias Hein

University of Tübingen, Germany

Stefanie Jegelka

Massachusetts Institute of Technology, USA

Arnulf Jentzen

Universität Münster, Germany

Kathlén Kohn

KTH - Royal Institute of Technology, Sweden

Věra Kůrková

Czech Academy of Sciences, Czech Republic

Jason Lee

Princeton University, USA

Stanley Osher

University of California, Los Angeles, USA

Jeffrey Pennington

Google Brain, USA

Stefano Soatto

University of California, Los Angeles, USA

Ingo Steinwart

University of Stuttgart, Germany

Dmitry Yarotsky

Skoltech, Russia

Lenka Zdeborová

EPFL - École Polytechnique Fédérale de Lausanne, Switzerland


10:00 - 10:30
10:30 - 11:00
11:00 - 12:00 Lenka Zdeborová (EPFL - École Polytechnique Fédérale de Lausanne, Switzerland)
12:00 - 12:30 Youness Boutaib (RWTH Aachen University, Aachen, Germany)
Path classification by stochastic linear RNNs
12:30 - 13:00 Yuguang Wang (Max Planck Institute for Mathematics in the Sciences, Leipzig, Germany)
How framelets enhance graph neural networks
13:00 - 14:00
14:00 - 14:30 Luca Ratti (University of Genoa, Genova, Italy)
Learning the optimal regularizer for linear inverse problems
14:30 - 15:00 Florent Krzakala (EPFL, Lausanne, Switzerland)
Generalization & Overparametrization in Machine Learning: Rigorous Insights from Simple Models
15:00 - 15:30
15:30 - 16:00 Oxana Manita (Eindhoven University of Technology, Eindhoven, Netherlands)
Dropout regularization viewed from the large deviations perspective
16:00 - 16:30 Jochen Merker (HTWK Leipzig, Leipzig, Germany)
16:30 - 17:00
17:00 - 18:00 Peter Bartlett (University of California, Berkeley, USA)
Benign overfitting
18:00 - 19:00 Stefano Soatto (University of California, Los Angeles, USA)
The Information in Optimal Representations
19:00 - 00:00
09:30 - 10:30 Ingo Steinwart (University of Stuttgart, Germany)
10:30 - 11:00
11:00 - 12:00 Dmitry Yarotsky (Skoltech, Russia)
12:00 - 12:30 Andreas Habring (University of Graz, Graz, Austria)
A Generative Variational Model for Inverse Problems in Imaging
12:30 - 14:00
14:00 - 15:00
Poster Session
15:00 - 15:30
15:30 - 16:30 Stefanie Jegelka (Massachusetts Institute of Technology, USA)
Learning and Generalization in Graph Neural Networks
16:30 - 17:00
17:00 - 18:00 Pierre Baldi (University of California, Irvine, USA)
18:00 - 19:00 Yasaman Bahri (Google Brain, USA)
Dynamics & Phase Transitions in Wide, Deep Neural Networks
09:30 - 10:30 Matthias Hein (University of Tübingen, Germany)
Towards neural networks which know when they don't know
10:30 - 11:00
11:00 - 12:00 Kathlén Kohn (KTH - Royal Institute of Technology, Sweden)
The Geometry of Linear Convolutional Networks
12:00 - 12:30 Luigi Malagò (Transylvanian Institute of Neuroscience (TINS), Cluj-Napoca, Romania)
Lagrangian and Hamiltonian Mechanics for Probabilities on the Statistical Bundle
12:30 - 13:00 Kirandeep Kour (Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany)
A Low-rank Support Tensor Network
13:00 - 14:00
14:00 - 14:30 Matthias Löffler (ETH Zürich, Zurich, Switzerland)
AdaBoost and robust one-bit compressed sensing
14:30 - 15:00 Benjamin Fehrman (University of Oxford, Oxford, United Kingdom)
15:00 - 15:30
15:30 - 16:30 Terry Lyons (University of Oxford, United Kingdom)
16:30 - 17:00
17:00 - 17:30 Michael Murray (University of Oxford, Oxford, UK, United Kingdom)
17:30 - 18:00 Daniel McKenzie (University of California, Los Angeles, Los Angeles, USA)
18:00 - 19:00 Jeffrey Pennington (Google Brain, USA)
19:00 - 00:00
09:30 - 10:30 Arnulf Jentzen (Universität Münster, Germany)
10:30 - 11:00
11:00 - 12:00 Minh Hà Quang (RIKEN-AIP, Japan)
12:00 - 12:30 Hanyuan Hang (University of Twente, Enschede, Netherlands)
A Combination of Ensemble Methods for Large-Scale Regression
12:30 - 13:00 Soon Hoe Lim (Nordita, KTH Royal Institute of Technology and Stockholm University, Stockholm, Sweden)
Noisy Recurrent Neural Networks
13:00 - 14:00
14:00 - 14:30 Guido Montúfar (Max Planck Institute for Mathematics in the Sciences, Leipzig, Germany)
Implicit bias of gradient descent for mean squared error regression with wide neural networks
14:30 - 15:00 Michael Schmischke (Chemnitz University of Technology, Chemnitz, Germany)
High-Dimensional Explainable ANOVA Approximation
15:00 - 15:30
15:30 - 16:00 Sebastian Kassing (WWU Münster, Münster, Germany)
16:00 - 16:30 Burim Ramosaj (TU Dortmund, Dortmund, Germany)
16:30 - 17:00
17:00 - 18:00 Jason Lee (Princeton University, USA)
18:00 - 19:00 Věra Kůrková (Czech Academy of Sciences, Czech Republic)

Scientific Organizers

Benjamin Gess

Max Planck Institute for Mathematics in the Sciences and Universität Bielefeld

Guido Montúfar

Max Planck Institute for Mathematics in the Sciences and UCLA

Nihat Ay

Hamburg University of Technology