Over the last years deep learning has developed into one of the most important areas of machine learning leading to break throughs in various applied fields like image and natural language processing or machine translation. These numerous advances on practical side are accompanied by a rather limited but growing theoretical understanding. Important questions relating to the representational power of the models, the interpretability of the solutions obtained, the stability and understanding of the stochastic optimization process, the generalization performance of deep neural networks, and new mathematical frameworks to learn generative models - just to name some - require us to delve deeper into the mathematics underlying the field of . deep learning. In this workshop we will discuss recent achievements, status quo, and open questions regarding our theoretical understanding of deep learning.
11:00- 11:30 Coffee 11:30- 12:00
13:00- 14:00 Lunch 18:00- 18:30