Jan Gerken

Postdoctor, Mathematical Sciences

I’m a postdoctoral researcher in the group of Daniel Persson working on deep learning, more specifically on group equivariant convolutional neural networks in the broader field of geometric deep learning. On the theoretical side, our work is heavily inspired by physics, using differential geometry and representation theory to construct and understand neural networks. On the applied side, we work together with Zenuity to improve object detection in self-driving cars.

Published: Mon 05 Oct 2020.