Jan Gerken

Assistant Professor at Algebra and geometry

I lead a WASP-funded group at the division of Algebra and Geometry in the area of mathematical foundations of deep learning. Our main focus lies on equivariant- and wide neural networks. On the theory side, our work is heavily inspired by physics, borrowing from concepts like gauge theory, general relativity and quantum field theory. On the practical side, we work on applications in condensed matter physics, quantum chemistry and computer vision.

For more information, have a look at our group homepage linked above.

Source: chalmers.se
Image of Jan Gerken

Showing 5 publications

2024

Diffeomorphic Counterfactuals with Generative Models

Ann Kathrin Dombrowski, Jan Gerken, Klaus Robert Muller et al
IEEE Transactions on Pattern Analysis and Machine Intelligence. Vol. 46 (5), p. 3257-3274
Journal article
2024

Emergent Equivariance in Deep Ensembles

Jan Gerken, Pan Kessel
Proceedings of the 38th Conference on Uncertainty in Artificial Intelligence, UAI 2022. Vol. 235, p. 15438-15465
Paper in proceeding
2024

HEAL-SWIN: A Vision Transformer on the Sphere

Oscar Carlsson, Jan Gerken, Hampus Linander et al
Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, p. 6067-6077
Paper in proceeding
2022

Equivariance versus Augmentation for Spherical Images

Jan Gerken, Oscar Carlsson, Hampus Linander et al
Proceedings of Machine Learning Research. Vol. 162, p. 7404-7421
Paper in proceeding
2022

Towards closed strings as single-valued open strings at genus one

Jan Gerken, Axel Kleinschmidt, Carlos R. Mafra et al
Journal of Physics A: Mathematical and Theoretical. Vol. 55 (2)
Journal article

Download publication list

You can download this list to your computer.

Filter and download publication list

As logged in user (Chalmers employee) you find more export functions in MyResearch.

You may also import these directly to Zotero or Mendeley by using a browser plugin. These are found herer:

Zotero Connector
Mendeley Web Importer

The service SwePub offers export of contents from Research in other formats, such as Harvard and Oxford in .RIS, BibTex and RefWorks format.

There are no projects.
There might be more projects where Jan Gerken participates, but you have to be logged in as a Chalmers employee to see them.