Matthias Kellner

Lausanne, Switzerland

Hi there! Welcome to my personal webpage!

I am Matthias, a fourth-year PhD student in computational materials science. My research interests include uncertainty quantification for atomistic modeling, predicting chemical shifts, and the development of universal machine learning force fields.

Currently I am based in Lausanne, Switzerland, working in the COSMO lab at EPFL under the supervision of Michele Ceriotti. I plan to graduate in 2027. Previously I obtained my Bachelor’s at TU Darmstadt and my Master’s from TU Munich.

When I am not researching I enjoy playing music (I play trombone and guitar), sailing, running and hiking!

If you are interested in my work or just want to chat, feel free to reach out on the socials below or drop me an email at matthias.kellner [at] epfl [dot] ch!

Try my work on the web

shiftml.org hosts ShiftML4 online, the latest chemical shielding model I jointly developed with the LRM laboratory at EPFL. No installation needed, it runs directly in your browser.

Open shiftml.org

news

Sep 15, 2026	I am giving an MLIP workshop together with Philip Loche at the University of Bergen!
Aug 21, 2026	We released ShiftML4, a fast and accurate chemical shielding predictor! Read the preprint on arXiv or try the model directly in your browser at shiftml.org.

selected publications

arXiv

Machine-Learned NMR Shieldings in Molecular Solids with Built-In Hybrid-Functional Molecular Corrections

Matthias Kellner, Ruben Rodriguez-Madrid, Jacob B Holmes, and 4 more authors

arXiv preprint, 2026

arXiv Bib Code Website

@article{kellner2026machine,
  title = {Machine-Learned NMR Shieldings in Molecular Solids with Built-In Hybrid-Functional Molecular Corrections},
  author = {Kellner, Matthias and Rodriguez-Madrid, Ruben and Holmes, Jacob B and Unzueta, Pablo A and Beran, Gregory JO and Emsley, Lyndon and Ceriotti, Michele},
  journal = {arXiv preprint},
  year = {2026},
}

arXiv

Errors that matter: Uncertainty-aware universal machine-learning potentials calibrated on experiments

Matthias Kellner, Teitur Hansen, Thomas Bligaard, and 2 more authors

arXiv preprint, 2026

arXiv Bib

@article{kellner2026errors,
  title = {Errors that matter: Uncertainty-aware universal machine-learning potentials calibrated on experiments},
  author = {Kellner, Matthias and Hansen, Teitur and Bligaard, Thomas and Jacobsen, Karsten Wedel and Ceriotti, Michele},
  journal = {arXiv preprint},
  year = {2026},
}

arXiv

Quantum-corrected NMR crystallography at scale

Matthias Kellner, Ruben Rodriguez-Madrid, Jacob B Holmes, and 3 more authors

arXiv preprint, 2026

arXiv Bib

@article{kellner2026quantum,
  title = {Quantum-corrected NMR crystallography at scale},
  author = {Kellner, Matthias and Rodriguez-Madrid, Ruben and Holmes, Jacob B and Principe, Victor Paul and Emsley, Lyndon and Ceriotti, Michele},
  journal = {arXiv preprint},
  year = {2026},
}

Nat. Commun.

PET-MAD as a lightweight universal interatomic potential for advanced materials modeling

Arslan Mazitov, Filippo Bigi, Matthias Kellner, and 6 more authors

Nature Communications, 2025

DOI Bib Code

@article{mazitov2025pet,
  title = {PET-MAD as a lightweight universal interatomic potential for advanced materials modeling},
  author = {Mazitov, Arslan and Bigi, Filippo and Kellner, Matthias and Pegolo, Paolo and Tisi, Davide and Fraux, Guillaume and Pozdnyakov, Sergey and Loche, Philip and Ceriotti, Michele},
  journal = {Nature Communications},
  volume = {16},
  number = {1},
  pages = {10653},
  year = {2025},
  publisher = {Nature Publishing Group},
  doi = {10.1038/s41467-025-65662-7},
}

JPCL

A deep learning model for chemical shieldings in molecular organic solids including anisotropy

Matthias Kellner, Jacob B Holmes, Ruben Rodriguez-Madrid, and 4 more authors

The Journal of Physical Chemistry Letters, 2025

DOI Bib Code

@article{kellner2025deep,
  title = {A deep learning model for chemical shieldings in molecular organic solids including anisotropy},
  author = {Kellner, Matthias and Holmes, Jacob B and Rodriguez-Madrid, Ruben and Viscosi, Florian and Zhang, Yuxuan and Emsley, Lyndon and Ceriotti, Michele},
  journal = {The Journal of Physical Chemistry Letters},
  volume = {16},
  number = {34},
  pages = {8714--8722},
  year = {2025},
  publisher = {American Chemical Society},
  doi = {10.1021/acs.jpclett.5c01819},
}

MLST

Uncertainty quantification by direct propagation of shallow ensembles

Matthias Kellner and Michele Ceriotti

Machine Learning: Science and Technology, 2024

DOI Bib Code

@article{kellner2024uncertainty,
  title = {Uncertainty quantification by direct propagation of shallow ensembles},
  author = {Kellner, Matthias and Ceriotti, Michele},
  journal = {Machine Learning: Science and Technology},
  volume = {5},
  number = {3},
  pages = {035006},
  year = {2024},
  publisher = {IOP Publishing},
  doi = {10.1088/2632-2153/ad594a},
}