About me

I am a PhD student at Humboldt-Universität zu Berlin with research interests in mathematical image processing, inverse problems and machine learning. I work within the MATH+ project AA5-6 under the supervision of Gabriele Steidl (TU Berlin) and Andrea Walther (HU Berlin). I obtained my BSc and MSc degrees in Mathematics in 2019 and 2021 from TU Berlin.

Publications

Stability of Data-Dependent Ridge-Regularization for Inverse Problems.
S. Neumayer and F. Altekrüger (2024).
(arXiv Preprint#2406.12289).
[arxiv], [code]

Posterior Sampling Based on Gradient Flows of the MMD with Negative Distance Kernel.
P. Hagemann, J. Hertrich, F. Altekrüger, R. Beinert, J. Chemseddine and G. Steidl (2024).
International Conference on Learning Representations 2024.
[openreview], [arxiv], [code], [pitch]

Generative Sliced MMD flows with Riesz kernels.
J. Hertrich, C. Wald, F. Altekrüger and P. Hagemann (2024).
International Conference on Learning Representations 2024.
[openreview], [arxiv], [code], [pitch]

Learning from small data sets: Patch-based regularizers in inverse problems for image reconstruction.
M. Piening, F. Altekrüger, J. Hertrich, P. Hagemann, A. Walther and G. Steidl (2024).
GAMM-Mitteilungen.
[doi], [arxiv], [code]

Neural Wasserstein Gradient Flows for Maximum Mean Discrepancies with Riesz Kernels.
F. Altekrüger, J. Hertrich and G. Steidl (2023).
International Conference on Machine Learning 2023.
Proceedings of Machine Learning Research, vol. 202, pp. 664-690.
[www], [arxiv], [code], [pitch]

Conditional Generative Models are Provably Robust: Pointwise Guarantees for Bayesian Inverse Problems.
F. Altekrüger, P. Hagemann and G. Steidl (2023).
Transactions on Machine Learning Research (TMLR).
[openreview]

Learning Regularization Parameter-Maps for Variational Image Reconstruction using Deep Neural Networks and Algorithm Unrolling.
A. Kofler, F. Altekrüger, F. A. Ba, C. Kolbitsch, E. Papoutsellis, D. Schote, C. Sirotenko, F. Zimmermann and K. Papafitsoros (2023).
SIAM Journal on Imaging Sciences, vol. 16(4), pp. 2202-2246.
[doi], [arxiv], [code]

PatchNR: Learning from Very Few Images by Patch Normalizing Flow Regularization.
F. Altekrüger, A. Denker, P. Hagemann, J. Hertrich, P. Maass and G. Steidl (2023).
Inverse Problems, vol. 39, no. 6.
[doi], [arxiv], [code]

WPPNets and WPPFlows: The Power of Wasserstein Patch Priors for Superresolution.
F. Altekrüger and J. Hertrich (2023).
SIAM Journal on Imaging Sciences, vol. 16(3), pp. 1033-1067.
[doi], [arxiv], [code]

Theses

Solving Bayesian Inverse Problems using Neural Networks.
F. Altekrüger (2024).
PhD Thesis, TU Berlin.
[doi]