WIAS invites applications for a
PhD student position (f/m/d)
(Ref. 21/16)

in the Research Group
”Nonlinear Optimization and Inverse Problems”
(Head: Prof. Dietmar Hömberg, PI: Dr. Martin Eigel) starting October 1, 2021.
The candidate is required to have a completed scientific university education (by the starting date of the position) in mathematics or a closely related field. Moreover, demonstrable programming experience preferrably in python and good communication skills in English are expected.
The successful candidate will work on the project „Adaptive Tensor Neural Networks for parametric PDEs“ within the DFG Priority Programme SPP 2298 „Theoretical Foundations of Deep Learning“. The goal of the project is the development of adaptive methods for neural networks and hybrid tensor neural networks for the solution of high-dimensional parametric PDEs. It is part of a collaboration between WIAS and the group of Prof. Lars Grasedyck at the RWTH Aachen.
The applicants are expected to have a strong mathematical background and to be familiar with some of the topics numerical analysis (for differential equations), uncertainty quantification, statistical learning theory, high-dimensional approximations (in particular low-rank tensor networks). Moreover, the applicants are expected to be experienced with at least one of the common python machine learning frameworks.
Queries about the project can be directed to Dr. Martin Eigel (Martin.Eigel@wias-berlin.de).

See here for more information: https://short.sg/j/11637472