Cancelled - Inaugural lecture Francqui Chair Prof. Dr. Gitta Kutyniok
Postponed to fall due to Corona measures
With great pleasure we announce that Professor Dr. Gitta Kutyniok will take up the Francqui Chair at the Department of Mathematics and Data Science (WIDS), research group Digital Mathematics (DIMA), VUB.
Prof. Dr. Gitta Kutyniok is renowned for her mathematical approaches to solve problems from data science such as inverse problems in imaging science (feature extraction, inpainting, etc.) or analysis and classification of high-dimensional data. She is Einstein Professor of Mathematics and a professor of computer science and electrical engineering at the Technical University of Berlin. She exploits methodologies from the areas of applied harmonic analysis, approximation theory, compressed sensing, frame theory, and functional analysis. Possible applications include medicine such as magnetic resonance imaging and analysis of proteomics data as well as telecommunication such as massive MIMO. She was the Emmy-Noether Lecturer of the German Mathematical Society in 2013 and became a member of the Berlin-Brandenburg Academy of Sciences and Humanities in 2016. In 2019 she was named a SIAM Fellow "for contributions to applied harmonic analysis, compressed sensing, and imaging sciences". She has been selected as a plenary speaker at the eighth European Congress of Mathematics in 2020.
Deep Learning meets Modeling: Taking the Best out of Both Worlds
When: Thursday 23 April 2020 at 18:00. The inaugural lecture will be followed by a reception.
Where: Building I, Room I.0.03, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels
Abstract: Pure model-based approaches are today often insufficient for solving complex inverse problems in imaging. At the same time, we witness the tremendous success of data-based methodologies, in particular, deep neural networks for such problems. However, pure deep learning approaches often neglect known and valuable information from the modeling world.
In this talk, we will provide an introduction to this complex problem and then focus on the inverse problem of (limited-angle) computed tomography. We will develop a conceptual approach by combining the model-based method of sparse regularization by shearlets with the data-driven method of deep learning. Our solvers are guided by a microlocal analysis viewpoint to pay particular attention to the singularity structures of the data. Finally, we will show that our algorithm significantly outperforms previous methodologies, including methods entirely based on deep learning.
The inaugural lecture will be followed by four course lectures taking place at the following dates:
Friday 24 April 2020, 13:30 – 16:00, room I.0.01:
Shearlets - From Theory to Applications
Monday 27 April 2020, 13:30 – 16:00, room I.2.02:
Mathematical Foundations of Deep Neural Networks
Wednesday 29 April 2020, 13:30 – 16:00, room I.0.03:
Deep Learning, Inverse Problems, and Sparse Regularization
Thursday 30 April 2020, 13:30 – 16:00, room I.0.01:
Deep Learning meets Partial Differential Equations
The course will be introductory and accessible to all with a basic knowledge of functional analysis.
For questions or late registration, please contact:
Sponsored by the Francqui Foundation.