Martin Burger
DESY and Professor of Mathematics at the University of Hamburg, Germany
DESY and Professor of Mathematics at the University of Hamburg, Germany
Bio: Martin Burger is a leading scientist at DESY and Professor of Mathematics at the University of Hamburg. He heads the Computional Imaging working group at DESY, which is also a research unit of the Helmholtz Imaging cooperation platform. Together with his team, he develops imaging reconstruction methods and novel mathematical tools with the aim of improving the reconstruction and analysis of scientific images. Moreover, they are concerned with the mathematical foundations of machine learning and of algorithms in data science.
Martin has a background in applied mathematics, specialising in inverse problems, mathematical modelling and partial differential equations. He is a world-leading expert in inverse problems and applied mathematics. He has received numerous awards and was invited speaker at all major international conferences. Before joining DESY and Helmholtz Imaging, he was a professor at the University of Münster and later at the Friedrich-Alexander University Erlangen-Nürnberg.
KEYNOTE
Robustness Issues in Machine Learning
Abstract: In this talk we will discuss some (missing) robustness issues in machine learning, in particular when applied to the solution of inverse problems. As particular instances we will highlight the potential appearance of hallucinations in supervised learning and adversarial examples. Finally we will discuss several strategies to improve robustness in a computationally tractable way.