|
I am a PhD student at Imperial College London, co-supervised by Dr. Björn Schuller and Dr. Tolga Birdal. My research is on geometric and topological deep learning and on generative models — spectral operators for graphs and higher-order structures, and diffusion and flow matching for discrete and scientific domains, with applications from quantum computing to imaging. I am currently a research intern at Quantinuum, working on quantum error correction. Feel free to drop me an email if you are interested. |
|
|
|
|
|
|
|
Krahn M*, Bastian L.*, Garg V., Schuller B., Birdal T. (* denotes shared first author) ICML, 2026 project page A vertex-level spectral operator that condenses higher-order topology via Schur complementation of a graded Laplacian |
|
Krahn M, Sasdelli M., Yang F, Golyanik V., Kannala J., Chin T., Birdal T. BMVC, 2024 project page A quantum annealer deployable optimiser for binary neural networks |
|
Debner A., Krahn M, Hirvisalo V. NLDL, 2024 code A reinforcement learning based scheduler for conditional task graphs |
|
Behmanesh M*, Krahn M*, Ovsjanikov M. (* denotes shared first author) ICML, 2023 code A new graph neural network architecture based on time derivative diffusion for node classification |
|
Benkner, M. S., Krahn M., Tretschk, E., Lähner, Z., Moeller, M. , Golyanik V. ICLR, 2023 as spotlight (top 20%) project website A deep learning framework to learn QUBO representations for computer vision problems |
|
Barlas P., Krahn M., Kleanthous S., Kyriakou K., Otterbacher J. ICWSM, 2022 A study of biases of commercially available image tagging services |
|
Krahn M, Bernard F., Golyanik V. 3DV, 2021 project page A SDP formulation for joint graph matching and clustering |
|
|
|