Welcome!
The number of successful applications of deep learning in the context of inverse problems rapidly grows. In this autumn school we lay the foundations of both machine learning and inverse problems, as well as cover current research topics in their intersection. The autumn school targets advanced Master students, PhD students and PostDocs working on inverse problems, who are interested in the intersection of deep learning and inverse problems.
Lecturers
Carola-Bibiane Schönlieb (University of Cambridge)Markus Haltmeier (University of Innsbruck)
Martin Benning (Queen Mary University of London)
Matthias Bethge (Max Planck Institute, Tübingen)
Michael Möller (University of Siegen)
Nihat Ay (Max Planck Institute, Leipzig)
Ozan Öktem (KTH Stockholm)
Simon Arridge (University College London)
Program
Registration will start on Monday at 8:15.
|
|
Monday 04.11. |
Tuesday 05.11. |
Wednesday 06.11. |
Thursday 07.11. |
Friday 08.11. |
| 09:00 - 10:30 |
Nihat Ay |
Matthias Bethge |
Markus Haltmeier |
Carola-Bibiane Schönlieb |
Simon Arridge |
| 10:30 - 11:00 |
Coffee |
Coffee |
Coffee |
Coffee |
Coffee |
| 11:00 - 12:30 |
Nihat Ay |
Matthias Bethge |
Ozan Öktem |
Simon Arridge |
Martin Benning |
| 12:30 - 14:00 |
Lunch |
Lunch |
Lunch |
Lunch |
Lunch |
| 14:00 - 15:30 |
Michael Möller |
Markus Haltmeier |
Ozan Öktem |
|
Martin Benning |
| 15:30 - 16:00 |
Coffee |
Coffee Poster Session |
Coffee |
Closing |
|
| 16:00 - 17:30 |
Michael Möller |
Carola-Bibiane Schönlieb |
|
||
| |
|
|
|
|
|
| Evening |
|
19:00 Night-watchman tour |
19:30 Conference Dinner |
|
|
Lectures
| Speaker |
Topic |
| Nihat Ay |
Artificial Neural Networks and Machine Learning: Theoretical Foundations |
| Michael Möller |
Optimization of Deep Neural Networks |
| Matthias Bethge |
1) Robust Decision Making
2) Adversarial Robustness |
| Markus Haltmeier |
1) Regularization of Inverse Problems and Null Space Networks 2) Data driven regularizers for Inverse Problems |
| Ozan Öktem |
Bayesian Inversion and Deep Learning |
| Carola-Bibiane Schönlieb |
Learning of regularization for Inverse Problems |
| Simon Arridge |
1) Combining learned and model-based approaches for Inverse Problems 2) Learned PDE methods for forward and Inverse Problems |
| Martin Benning |
1) Iterative methods in Inverse Problems and Machine Learning 2) Nonlinear spectral transformation in Inverse Problems and Machine Learning |
Organization committee
Christian Etmann (University of Bremen, Germany)Daniel Otero Baguer (University of Bremen, Germany)
Jens Behrmann (University of Bremen, Germany)
Sören Dittmer (University of Bremen, Germany)
Dr. Tobias Kluth (University of Bremen, Germany)
Prof. Dr. Dr. h.c. Peter Maass (University of Bremen, Germany)
Acknowledgments
We gratefully acknowledge the support of this summer school by the DFG Research Training Group 2224 on Parameter Identification at the Center for Industrial Mathematics, University of Bremen.Thanks to Matthias Knauer for his help in setting up this website.
