Interests

Bayesian statistics
Bayesian inverse problems — Gaussian Process Regression

Uncertainty Quantification
Model parameter estimation — Sensitivity analysis — Informed decisions

Subspace-based dimension reduction
Active subspaces

Publications

Preprints

–

2023

TP., M., Brandl, G., Franz, C., Stuhr, U., Ganeva, M., & Schneidewind, A. (2023). Active learning-assisted neutron spectroscopy with log-Gaussian processes. Nature Communications 14, 2246. doi:10.1038/s41467-023-37418-8 – journal, arXiv

2022

TP., M., Schneidewind, A., Brandl, G., Franz, C., Noack, M., Boehm, M., & Ganeva, M. (2022). Benchmarking autonomous scattering experiments illustrated on TAS. Frontiers in Materials 8, 772014. doi:10.3389/fmats.2021.772014 – journal

2020

PhD thesis. Active Subspaces in Bayesian Inverse Problems. Technical University Munich – link

TP., M., Wallin, J., & Wohlmuth, B. (2020). Generalized bounds for active subspaces. Electronic Journal of Statistics 14(1), 917–943. doi:10.1214/20-EJS1684 – journal, arXiv

Bittner, D., TP., M., Mattis, S., Wohlmuth, B., & Chiogna, G. (2020). Identifying relevant hydrological and catchment properties in active subspaces: An inference study of a lumped karst aquifer model. Advances in Water Resources 135, 103472. doi:10.1016/j.advwatres.2019.103472 – journal

2019

TP., M., Bittner, D., Mattis, S., Chiogna, G., & Wohlmuth, B. (2019). Bayesian calibration and sensitivity analysis for a karst aquifer model using active subspaces. Water Resources Research 55(8), 7086–7107. doi:10.1029/2019WR024739 – journal, arXiv

TP., M., Mattis, S., Gupta, S., Deusner, C., & Wohlmuth, B. (2019). Efficient parameter estimation for a methane hydrate model with active subspaces. Computational Geosciences 23(2), 355–372. doi:10.1007/s10596-018-9769-x – journal (full-text view-only), arXiv

2016

Master thesis. Brownian Motion and the Dirichlet Problem. Ludwig-Maximilians-Universität München (LMU) – pdf

2013

Bachelor thesis. N.V. Krylov’s Proof of the de Moivre-Laplace Theorem. University of Applied Sciences Munich (HM) – pdf

Not published

TP., M. (2018). A probabilistic framework for approximating functions in active subspaces – arXiv

Talks, Conferences, etc.

Helmholtz AI Conference 2023. Active learning-assisted neutron spectroscopy with log-Gaussian processes. Helmholtz AI, Helmholtz Association. June 2023

Machine Learning Workshop. Active learning-assisted neutron spectroscopy with log-Gaussian processes. Lawrence Berkeley National Laboratory. April 2023

ECNS 2023. AI-assisted neutron spectroscopy – Log-Gaussian processes for TAS. Heinz Maier-Leibnitz Zentrum. March 2023

JCNS Workshop (invited talk). AI-assisted neutron spectroscopy – Log-Gaussian processes for TAS. Jülich Centre for Neutron Science. October 2022

MLZ User Meeting. Benchmarking autonomous TAS experiments. Heinz Maier-Leibnitz Zentrum. December 2021

Workshop on SAXS@XFELs and HI & HE laser driven matter. Benchmarking autonomous TAS experiments. Helmholtz-Zentrum Dresden-Rossendorf. November 2021

Workshop on Innovative Inelastic Neutron Scattering. Benchmarking autonomous scattering experiments illustrated on TAS. Institut Laue-Langevin. October 2021

Workshop on Autonomous Discovery in Science and Engineering (website). Autonomous Experiments for Neutron Three-Axis Spectrometers (TAS) with Log-Gaussian Processes. Center for Advanced Mathematics for Energy Research Applications, Lawrence Berkeley National Laboratory. April 2021 – extended abstract, post

SIAM UQ 2020. Solving a Bayesian Inverse Problem for a Karst Aquifer Model with Active Subspaces. Garching. March 2020 (canceled due to outbreak of SARS-CoV-2)

Statistics seminar. Active Subspaces in Bayesian Inverse Problems. Department of Statistics, Lund University. May 2019

M2 Oberseminar. Active subspaces for Bayesian inversion, Application to a methane hydrate model. Garching. March 2018 – post, slides

Awards / Prizes

Helmholtz AI Award. Best Paper 2023. December 2023

Press / Media

JCNS News. “Best Paper Award” from Helmholtz AI. February 2024 – link

Helmholtz AI News. Helmholtz AI Awards 2023: Meet Our Awardees. December 2023 – link

MLZ Newsletter. Machine learning tool assists TAS users. August 2023 – link

Nature Portfolio Instagram. Artificial intelligence speeds up material research. July 2023 – link

LENS Science Highlight. Efficient Use of Measurement Time Through Machine Learning. June 2023 – link

MLZ/FRM2 Press release. Using measurement time more efficiently with machine learning. June 2023 – link MLZ/link FRM2

FZJ Scientific Highlight. Neutron Research: Efficient Use of Measurement Time Through Machine Learning. May 2023 – link

“Behind the Paper” blog post. Towards AI-assisted neutron spectroscopy. April 2023 – link