Leibniz MMS Summer School 2021

August 23, 2021 - August 27, 2021

Venue: Schloss Dagstuhl - Leibniz-Zentrum für Informatik GmbH

In the last years, modern Machine Learning (ML) and in particular Deep Learning (DL) methods, have revolutionized many application areas (e.g. image recognition, natural language processing, etc.) in research and industry. These methods have also started to be used in the field of scientific computing to speed up simulations, interpret measurement and simulation data and to recover dynamics.

In this school, fundamental and advanced aspects of ML methods were presented with the aim to enable the participants to apply such methods to their specific research problems. In particular, the addressed topics included

• Fundamentals of statistical learning theory
• Introduction to machine learning with python
• Signatures and learning of time-invariant features of time series
• Data driven modeling of dynamical systems
• Distributed learning
• Robust methods for machine learning and data-driven decision-making
• Kernel methods

The morning sessions were generally devoted to theoretical introductions to the topics mentioned above, which were then practiced in applied lab sessions prepared and supervised by the speakers in the afternoons. The goal of the summer school was to give the participants a practical entry into the vast field of machine learning underpinned by understanding of the underlying theoretical concepts.

In addition, the summer school provided opportunities for exchange between students and networking, especially important due to the limitations because of the COVID 19 pandemic. In fact, it was the first summer school or workshop with personal attendance for many of the participants.

Particular thanks go to the speakers

• Christian Bayer (WIAS, organizer)
• Martin Eigel (WIAS, organizer)
• Nicole Mücke (TU Braunschweig)
• Feliks Nüske (Paderborn University)
• David Sommer (WIAS Berlin)
• Nikolas Tapia (WIAS Berlin)
• Jia-Jie (JJ) Zhu (WIAS Berlin)

as well as the staff at Schloss Dagstuhl for making the summer school possible despite the many challenges due to the pandemic.

Image copyrights:

• header image and lower figure left (back propagation example) under license CC BY 4.0 (header image slightly fomally modified)
• drawing 1st paragraph: WIAS
• lower figure right (artificial neural network with chip) under license CC BY 2.0