Prof. Dr. Mathias Trabs
- Raum: 2.020
CS 20.30 - Tel.: +49 721 608-43709
- trabs ∂does-not-exist.kit edu
- Englerstraße 2
76131 Karlsruhe
Forschung
Forschungsinteressen
- Nichtparametrische und hochdimensionale Statistik
- Statistik für stochastische Prozesse
- Statistische inverse Probleme
- Statistisches Lernen
- Stochastische (partielle) Differentialgleichungen
Current Projects
DASHH is a Helmholtz graduate school involving several partner institutions in Hamburg. In DASHH we harness data, computer and applied mathematical science to advance our understanding of nature. We aim to educate the future generation of data- and information- scientists that will tackle tomorrow’s scientific challenges that come along with large-scale experiments.
Past Projects
- DFG project TR1349/3-1 "High-dimensional statistics for point and jump processes", GEPRIS (2019 - 2023)
- LD-SODA: Lernbasierte Datenanalyse – Stochastik, Optimierung, Dynamik und Approximation (Landesforschungsförderung Hamburg)
Current and past PhD students:
| 2022 - now | Lea Kunkel | Karlsruhe Institute of Technology | |
| 2021 - now | Thea Engler | Joint supervision with Christian Schroer and Johannes Hagemann | Desy |
| 2021 - 2025 | Jan Rabe | Asymptotic confidence bands for centered purely random forests (Joint supervision with Natalie Neumeyer) | Universität Hamburg |
| 2021 - 2024 | Sebasian Bieringer | Uncertainties in Generative Deep Learning and Data Amplification for High Energy Physics (joint supervision with Gregor Kasieczka) | Universität Hamburg |
| 2019 - 2024 | Maximilian F. Steffen | Multivariate estimation in nonparametric models: Stochastic neural networks and Lévy processes | Karlsruhe Institute of Technology |
| 2017 - 2021 | Florian Hildebrandt | Parameter estimation for SPDEs based on discrete observations in time and space | Universität Hamburg |
Short CV
| Since 2021 | Professor at Karlsruhe Institute of Technology |
| 2021 | Heisenberg professor at Universität Hamburg |
| 2016 - 2021 | Assistant professor at Universität Hamburg |
| 2015 - 2016 | DFG research fellow at Université Paris-Dauphine |
| 2014 - 2015 | Postdoctoral researcher at Humboldt-Universität zu Berlin |
| 2014 | Visiting Ph.D. student at the University of Cambridge, UK |
| 2011 - 2014 | Ph.D. study in Mathematics at Humboldt-Universität zu Berlin |
| 2007 - 2011 | Studies in Mathematics with minor Economics at Humboldt-Universität zu Berlin |
Bücher
-
Trabs, M.; Jirak, M.; Krenz, K.; Reiß, M.
Statistik und maschinelles Lernen – Eine mathematische Einführung in klassische und moderne Methoden
2021. Springer-Verlag. doi:10.1007/978-3-662-62938-3
Publikationen
-
Ehrenreich-Petersen, E.; Massani, B.; Engler, T.; Pardo, O. S.; Glazyrin, K.; Giordano, N.; Hagemann, J.; Sneed, D.; Fedotenko, T.; Campbell, D. J.; Wendt, M.; Wenz, S.; Schroer, C. G.; Trabs, M.; McWilliams, R. S.; Liermann, H.-P.; Jenei, Z.; O’Bannon, E. F.
X-ray phase contrast imaging and diffraction in the laser-heated diamond anvil cell: A case study on the high-pressure melting of Pt
2025. Results in Physics, 69, Art.-Nr.: 108132. doi:10.1016/j.rinp.2025.108132 -
Kunkel, L.; Trabs, M.
A Wasserstein perspective of Vanilla GANs
2025. Neural Networks, 181, 106770. doi:10.1016/j.neunet.2024.106770 -
Bieringer, S.; Diefenbacher, S.; Kasieczka, G.; Trabs, M.
Calibrating Bayesian generative machine learning for Bayesiamplification
2024. Machine Learning: Science and Technology, 5 (4), Art.-Nr.: 045044. doi:10.1088/2632-2153/ad9136 -
Bieringer, S.; Kasieczka, G.; Kieseler, J.; Trabs, M.
Classifier surrogates: sharing AI-based searches with the world
2024. The European Physical Journal C, 84 (9), Art.-Nr.: 972. doi:10.1140/epjc/s10052-024-13353-w -
Hildebrandt, F.; Trabs, M.
Nonparametric calibration for stochastic reaction–diffusion equations based on discrete observations
2023. Stochastic Processes and their Applications, 162, 171–217. doi:10.1016/j.spa.2023.04.019 -
Hoffmann, M.; Trabs, M.
Dispersal density estimation across scales
2023. The Annals of Statistics, 51 (3), 1258–1281. doi:10.1214/23-AOS2290 -
Eckstein, S.; Iske, A.; Trabs, M.
Dimensionality Reduction and Wasserstein Stability for Kernel Regression
2023. Journal of Machine Learning Research, 24 -
Bieringer, S.; Butter, A.; Diefenbacher, S.; Eren, E.; Gaede, F.; Hundhausen, D.; Kasieczka, G.; Nachman, B.; Plehn, T.; Trabs, M.
Calomplification — the power of generative calorimeter models
2022. Journal of Instrumentation, 17 (09), Art.Nr. P09028. doi:10.1088/1748-0221/17/09/P09028 -
Prömel, D. J.; Trabs, M.
Paracontrolled distribution approach to stochastic Volterra equations
2021. Journal of Differential Equations, 302, 222–272. doi:10.1016/j.jde.2021.08.031 -
Hildebrandt, F.; Trabs, M.
Parameter estimation for SPDEs based on discrete observations in time and space
2021. Electronic Journal of Statistics, 15 (1), 2716–2776. doi:10.1214/21-EJS1848 -
Trabs, N.; Trabs, M.; Stodieck, S.; House, P. M.
Influence of stiripentol on perampanel serum levels
2020. Epilepsy Research, 164, Article no: 106367. doi:10.1016/j.eplepsyres.2020.106367 -
Bibinger, M.; Trabs, M.
Volatility estimation for stochastic PDEs using high-frequency observations
2020. Stochastic Processes and their Applications, 130 (5), 3005–3052. doi:10.1016/j.spa.2019.09.002 -
Belomestny, D.; Trabs, M.; Tsybakov, A. B.
Sparse covariance matrix estimation in high-dimensional deconvolution
2019. Bernoulli, 25 (3). doi:10.3150/18-BEJ1040A -
Niebuhr, T.; Trabs, M.
Profiting from correlations: Adjusted estimators for categorical data
2019. Applied Stochastic Models in Business and Industry, 35 (4), 1090–1102. doi:10.1002/asmb.2452 -
Belomestny, D.; Trabs, M.
Low-rank diffusion matrix estimation for high-dimensional time-changed Lévy processes
2018. Annales de l’Institut Henri Poincaré, Probabilités et Statistiques, 54 (3), 1584–1621. doi:10.1214/17-AIHP849 -
Trabs, M.
Bayesian inverse problems with unknown operators
2018. Inverse Problems, 34 (8), Article no: 085001. doi:10.1088/1361-6420/aac3aa -
Chorowski, J.; Trabs, M.
Spectral estimation for diffusions with random sampling times
2016. Stochastic Processes and their Applications, 126 (10), 2976–3008. doi:10.1016/j.spa.2016.03.009 -
Prömel, D. J.; Trabs, M.
Rough differential equations driven by signals in Besov spaces
2016. Journal of Differential Equations, 260 (6), 5202–5249. doi:10.1016/j.jde.2015.12.012 -
Nickl, R.; Reiß, M.; Söhl, J.; Trabs, M.
High-frequency Donsker theorems for Lévy measures
2016. Probability Theory and Related Fields, 164 (1-2), 61–108. doi:10.1007/s00440-014-0607-3 -
Dattner, I.; Reiß, M.; Trabs, M.
Adaptive quantile estimation in deconvolution with unknown error distribution
2016. Bernoulli, 22 (1). doi:10.3150/14-BEJ626 -
Söhl, J.; Trabs, M.
Adaptive confidence bands for Markov chains and diffusions: Estimating the invariant measure and the drift
2016. ESAIM: Probability and Statistics, 20, 432–462. doi:10.1051/ps/2016017 -
Trabs, M.
Information bounds for inverse problems with application to deconvolution and Lévy models
2015. Annales de l’Institut Henri Poincaré, Probabilités et Statistiques, 51 (4), 1620–1650. doi:10.1214/14-AIHP627 -
Trabs, M.
Quantile estimation for Lévy measures
2015. Stochastic Processes and their Applications, 125 (9), 3484–3521. doi:10.1016/j.spa.2015.04.004 -
Söhl, J.; Trabs, M.
Option calibration of exponential Lévy models: confidence intervals and empirical results
2014. The Journal of Computational Finance, 18 (2), 91–119. doi:10.21314/JCF.2014.275 -
Trabs, M.
On infinitely divisible distributions with polynomially decaying characteristic functions
2014. Statistics & Probability Letters, 94, 56–62. doi:10.1016/j.spl.2014.07.002 -
Trabs, M.
Calibration of self-decomposable Lévy models
2014. Bernoulli, 20 (1). doi:10.3150/12-BEJ478 -
Söhl, J.; Trabs, M.
A uniform central limit theorem and efficiency for deconvolution estimators
2012. Electronic Journal of Statistics, 6, 2486–2518. doi:10.1214/12-EJS757
weitere Aktivitäten
- Deputy speaker of the KIT Center MathSEE Mathematics in Sciences, Engineering, and Economics
- Member of the steering board of the DMV-Fachgruppe Stochastik (Probability and Statistics Group of the German Mathematical Society)
- Member of the steering board of the KIT Graduate School Computational and Data Science (KCDS)
- Associate editor for Statistics