Technical University of Munich

Christian Koke

PhD StudentTechnical University of Munich

School of Computation, Information and Technology
Informatics 9
Boltzmannstrasse 3
85748 Garching
Germany

Tel: +49-89-289-17790
Fax: +49-89-289-17757
Office: 02.09.039
Mail: Christian.Koke@tum.de

Hi there!

I like Machine Learning, Physics and Math. If you would like to chat about these topics, feel free to write me an email!

Research Interests

I am currently working on graph neural networks, geometric deep learning and AI for science. I also remain interested in the mathematical foundations of machine learning.

About

I am an ELLIS PhD Student jointly supervised by Daniel Cremers here at TUM and Michael Bronstein at Oxford University. I hold M.Sc. degrees in Math and Theoretical Physics; the latter awarded within the joint Elite Graduate Program of LMU and TU Munich. Before joining the Computer Vision Group as a PhD Student, I worked with Gitta Kutyniok on the mathematical foundations of graph neural networks and spent a couple of summers interning at the Centre for Quantum Technologies in Singapore.

Publications

Google Scholar
Export as PDF, XML, TEX or BIB

2024

Conference and Workshop Papers
[]		Transferability for Graph Convolutional Networks (C Koke, A Saroha, Y Shen, M Eisenberger, MM. Bronstein and D Cremers), In ICML 2024 Workshop on Geometry-grounded Representation Learning and Generative Modeling, 2024. [bibtex] [pdf]Outstanding Extended Abstract
[]		HoloNets: Spectral Convolutions do extend to Directed Graphs (C Koke and D Cremers), In International Conference on Learning Representations (ICLR), 2024. [bibtex] [arXiv:2310.02232]

2023
Conference and Workshop Papers
[]		ResolvNet: A Graph Convolutional Network with multi-scale Consistency (C Koke, A Saroha, Y Shen, M Eisenberger and D Cremers), In NeurIPS 2023 Workshop: New Frontiers in Graph Learning, 2023. [bibtex] [arXiv:2310.00431]Oral Presentation
[]		HoloNets: Spectral Convolutions do extend to Directed Graphs (C Koke and D Cremers), In NeurIPS 2023 Workshop: New Frontiers in Graph Learning, 2023. [bibtex] [arXiv:2310.02232]Oral Presentation
[]		Limitless Stability for Graph Convolutional Networks (C Koke), In International Conference on Learning Representations (ICLR), 2023. [bibtex] [arXiv:2301.11443]

2022
Conference and Workshop Papers
[]		Graph Scattering beyond Wavelet Shackles (C Koke and G Kutyniok), In Neural Information Processing Systems Conference (NeurIPS), 2022. [bibtex] [arXiv:2301.11456]

2020
Journal Articles
[]		Dirac equation on a square waveguide lattice with site-dependent coupling strengths and the gravitational Aharonov-Bohm effect (C Koke, C Noh and D Angelakis), In Physical Review A, volume 102, 2020. [bibtex] [arXiv:1909.12543]

2016
Journal Articles
[]		Dirac equation in 2-dimensional curved spacetime, particle creation, and coupled waveguide arrays (C Koke, C Noh and D Angelakis), In Annals of Physics, volume 374, 2016. [bibtex] [arXiv:1607.04821]