Mathematik, Arbeitsrichtung Angewandte Mathematik

Paul Hagemann

Kontakt

Technische Universität Berlin Institut für Mathematik Sekretariat MA 4-3 Straße des 17. Juni 136 10623 Berlin
Raum MA 479 Tel.: +49 - (0)30 - 314-79758 Email: hagemann (at) math.tu-berlin.de

Sekretariat MA 4-3 Julia Wilton Raum MA 476 Email: wilton_(at) math.tu-berlin.de

 

Forschungsinteressen

Inverse Probleme, Generative Modelling und Uncertainty Quantification

Wissenschaftlicher Werdegang

10/2021: Wissenschaftlicher Mitarbeiter an der TU Berlin

Publikationen

F. Altekrüger, A. Denker, P. Hagemann, J. Hertrich, P. Maass and G. Steidl (2022).
PatchNR: Learning from Small Data by Patch Normalizing Flow Regularization.
(arXiv Preprint#2205.12021)
[arxiv][Code]

P. Hagemann, J. Hertrich and G. Steidl (2021).
Generalized Normalizing Flows via Markov Chains.
Accepted in: Elements in Non-local Data Interactions: Foundations and Applications.
Cambridge University Press.
[arxiv][Code]

P. Hagemann, J. Hertrich and G. Steidl (2022).
Stochastic Normalizing Flows: a Markov Chains Viewpoint.
SIAM/ASA Journal on Uncertainty Quantification, vol. 10, pp. 1162-1190.
[doi], [arxiv], [Code]

A. Andrle, N. Farchmin, P. Hagemann, S. Heidenreich, V. Soltwisch, G. Steidl (2021).
Invertible Neural Networks Versus MCMC for Posterior Reconstruction in Grazing Incidence X-Ray Fluorescence
Scale Space and Variational Methods in Computer Vision, 2021, Volume 12679                                                                                                                                                      [doi], [arxiv]

P. Hagemann, S. Neumayer (2021)
Stabilizing Invertible Neural Networks Using Mixture Models
Inverse Problems (Volume 37, Number 8)                                                                                                                                                                                                                          [doi], [arxiv], [Code]