The workshop is postponed due to the corona virus situation.
We are considering an alternative date towards the end of 2020.
Machine Learning in Imaging Sciences: Bridging the gap between theory and practice
Weierstrass Institute Berlin, 11-13 May 2020
Organizers
- Michael Hintermüller (WIAS Berlin, Humboldt University)
- Kostas Papafitsoros (WIAS Berlin)
- Guozhi Dong (WIAS Berlin, Humboldt University)
- Karsten Tabelow (WIAS Berlin)
- Tobias Schäffter (PTB)
- Gitta Kutyniok (TU Berlin)
- Hans-Christian Hege (ZIB)
Scope
Machine learning is widely used nowadays in imaging sciences delivering state-of-the-art results in many applications. The use of artificial neural networks along with the availability of large amounts of training data has led to impressive developments not only in traditional image reconstruction and restoration tasks but also to advanced imaging modalities. Of particular interest are recently developed imaging modalities in the field of medical imaging for which the image quality might be crucial for reliable diagnoses.
Yet, the use of neural networks in imaging sciences has spread too fast for the theory to follow. This means that for the majority of tasks, they are being used as black box with little understanding of their underlying properties. This is in strong contrast to the so far developed traditional techniques of (Mathematical) Imaging, where rigorous mathematics are able to provide reconstruction guarantees and understanding of the mechanisms of the methods and their corresponding results. It is thus, of immense importance to bridge the gap between theory and practice with respect to machine learning techniques in the field of imaging science.
The workshop aims to bring together scientists that are working on cutting edge applications of machine learning in imaging sciences as well as their mathematically theory.
Confirmed invited speakers
- Simon Arridge (University College London)
- Marta Betcke (University College London)
- Mike Davies (University of Edinburgh)
- Matthias Ehrhardt (University of Bath)
- Markus Haltmeier (University of Innsbruck)
- Anders Hansen (University of Cambridge)
- Martin Holler (University of Graz)
- Andreas Kofler (Charite)
- Andreas Maier (University of Erlangen)
- Jan Modersitzki (University of Lübeck)
- Ozan Öktem (KTH)
- Xiaoying Tang (Southern University of Science and Technology, Shenzhen)
- Lars Ruthotto (Emory University)
- Jing Yuan (Xidian University)
Further speakers to be confirmed soon.
Program
The detailed program of the workshop will be announced soon.
Registration
The registration is free but mandatory. Please note that due to the limited capacity of the venue (70 people) priority will be given to people that register first.
Please register following this link.
Contact
Email: ml-imaging2020@wias-berlin.de
Phone: +49 30 20372-382
Venue
The workshop will take place at the Weierstrass Institute (WIAS), which is located in a historic building in the center of Berlin.
- Address: Mohrenstraße 39, 10117 Berlin, Germany
(Subway line U2, station Hausvogteiplatz)
- You can find the location of WIAS
highlighted
here (Openstreetmap).
- Lecture room: ESH (ground floor)
- See www.bvg.de for more information about public transport in Berlin (routes/tickets).
Support
The workshop gratefully acknowledges the support of the Berlin Mathematics Research Centre MATH+, as well as of the DFG SPP 1962 Priority Programme on "Non-smooth and Complementarity-based Distributed Parameter Systems: Simulation and Hierarchical Optimization". The workshop is further supported by the Weierstrass Institute Berlin and the Humboldt University of Berlin.
Organizers
- Michael Hintermüller (WIAS Berlin, Humboldt University)
- Kostas Papafitsoros (WIAS Berlin)
- Guozhi Dong (WIAS Berlin, Humboldt University)
- Karsten Tabelow (WIAS Berlin)
- Tobias Schäffter (PTB)
- Gitta Kutyniok (TU Berlin)
- Hans-Christian Hege (ZIB)
Scope
Machine learning is widely used nowadays in imaging sciences delivering state-of-the-art results in many applications. The use of artificial neural networks along with the availability of large amounts of training data has led to impressive developments not only in traditional image reconstruction and restoration tasks but also to advanced imaging modalities. Of particular interest are recently developed imaging modalities in the field of medical imaging for which the image quality might be crucial for reliable diagnoses. Yet, the use of neural networks in imaging sciences has spread too fast for the theory to follow. This means that for the majority of tasks, they are being used as black box with little understanding of their underlying properties. This is in strong contrast to the so far developed traditional techniques of (Mathematical) Imaging, where rigorous mathematics are able to provide reconstruction guarantees and understanding of the mechanisms of the methods and their corresponding results. It is thus, of immense importance to bridge the gap between theory and practice with respect to machine learning techniques in the field of imaging science. The workshop aims to bring together scientists that are working on cutting edge applications of machine learning in imaging sciences as well as their mathematically theory.
Confirmed invited speakers
- Simon Arridge (University College London)
- Marta Betcke (University College London)
- Mike Davies (University of Edinburgh)
- Matthias Ehrhardt (University of Bath)
- Markus Haltmeier (University of Innsbruck)
- Anders Hansen (University of Cambridge)
- Martin Holler (University of Graz)
- Andreas Kofler (Charite)
- Andreas Maier (University of Erlangen)
- Jan Modersitzki (University of Lübeck)
- Ozan Öktem (KTH)
- Xiaoying Tang (Southern University of Science and Technology, Shenzhen)
- Lars Ruthotto (Emory University)
- Jing Yuan (Xidian University)
Further speakers to be confirmed soon.
Program
The detailed program of the workshop will be announced soon.
Registration
The registration is free but mandatory. Please note that due to the limited capacity of the venue (70 people) priority will be given to people that register first.
Please register following this link.Contact
Email: ml-imaging2020@wias-berlin.de
Phone: +49 30 20372-382
Venue
The workshop will take place at the Weierstrass Institute (WIAS), which is located in a historic building in the center of Berlin.
- Address: Mohrenstraße 39, 10117 Berlin, Germany (Subway line U2, station Hausvogteiplatz)
- You can find the location of WIAS highlighted here (Openstreetmap).
- Lecture room: ESH (ground floor)
- See www.bvg.de for more information about public transport in Berlin (routes/tickets).
Support
The workshop gratefully acknowledges the support of the Berlin Mathematics Research Centre MATH+, as well as of the DFG SPP 1962 Priority Programme on "Non-smooth and Complementarity-based Distributed Parameter Systems: Simulation and Hierarchical Optimization". The workshop is further supported by the Weierstrass Institute Berlin and the Humboldt University of Berlin.