Current Group Member

Prof. Dr. Roland Herzog

Group Leader

I joined Heidelberg University in April 2021 and established the group Scientific Computing and Optimization (SCOOP) at the Interdisciplinary Center for Scientific Computing (IWR).

You can find my CV here.

Research interests

My research interests include

  • optimal control of partial differential equations
  • large-scale optimization and its applications
  • numerical linear algebra
  • optimal experimental design
  • optimization on manifolds
  • numerical methods for partial differential equations

Current and Upcoming Projects

Computational Shape Optimization for Materials Science and Biology
Start: 2026-01-01
End: 2026-12-31
Principal Investigators: Roland Herzog, Ulrich Schwarz
Staff: Manuel Weiß
Funded by: MWK, BMFTR within the Excellence Strategy funding scheme
Part of: Patterns and Structures in Mathematics, Data, and the Material World (FoF 2)

Machine Learning and Optimal Experimental Design for Thermodynamic Property Modeling
Start: 2025-03-01
End: 2028-02-29
Principal Investigators: Roland Herzog, Markus Richter
Staff: Viktor Martinek, Ophelia Frotscher
Funded by: DFG within the Priority Program funding scheme
Part of: Machine Learning in Chemical Engineering (SPP 2331)

Surrogate-Based Prediction of Cycle Fatigue Strength of Rotating Shafts and Process Optimization Using Flexible Kriging Models (dPRO2)
Start: 2024-10-01
End: 2027-09-30
Principal Investigators: Roland Herzog
Collaborators: Thomas Lampke, Lisa Winter, Andreas Schubert
Staff: Lena Becker, Hannah Rickmann, Sahib Kaur, Pascale Neubauer
Funded by: DFG within the Package Proposal funding scheme
Part of: Data-Driven Process-Property Models to Predict Cycle Fatigue Strength (PAK 1105)

Optimal Control and Optimal Experimental Design for MRI and Photoacoustic Imaging (MBAI)
Start: 2022-04-01
End: 2028-03-31
Principal Investigators: Roland Herzog, Robert Scheichl
Staff: Karina Koval
Funded by: Carl Zeiss Foundation within the CZS Breakthroughs funding scheme
Part of: Model-Based AI: Physical Models and Deep Learning for Imaging and Cancer Treatment

Latest Publications

  • Roland Herzog and Viktor Martinek
    Machine learning for thermodynamic property modeling
    Oberwolfach Reports, 2026
    doi:10.4171/owr/2026/26
    bibtex 
    @ARTICLE{HerzogMartinek:2026:1,
  AUTHOR = {Herzog, Roland and Martinek, Viktor},
  EDITOR = {Ghattas, Omar and Heinkenschloss, Matthias and Reich, Sebastian and Schillings, Claudia},
  PUBLISHER = {European Mathematical Society Publishing House},
  BOOKTITLE = {Workshop: Mathematical Foundation of Digital Twins},
  DATE = {2026-08},
  DOI = {10.4171/owr/2026/26},
  JOURNALTITLE = {Oberwolfach Reports},
  TITLE = {Machine learning for thermodynamic property modeling},
}
  • Viktor Martinek and Roland Herzog
    Symbolic regression for shared expressions – introducing partial parameter sharing
    Parallel Problem Solving from Nature – PPSN XIX, p.251-266, 2026
    arXiv:2601.04051
    doi:10.1007/978-3-032-36229-2_16
    bibtex 
    @INPROCEEDINGS{MartinekHerzog:2026:2,
  AUTHOR = {Martinek, Viktor and Herzog, Roland},
  PUBLISHER = {Springer Nature Switzerland},
  BOOKTITLE = {Parallel Problem Solving from Nature – PPSN XIX},
  DATE = {2026-08},
  DOI = {10.1007/978-3-032-36229-2_16},
  EPRINT = {2601.04051},
  EPRINTTYPE = {arXiv},
  PAGES = {251--266},
  TITLE = {Symbolic regression for shared expressions – introducing partial parameter sharing},
}
  • Niklas Emonds, Evelyn Herberg, Martin Fungisai Gerchen, Marc Pritsch, Joshua Rocha, Vera Zamoscik, Peter Kirsch, Roland Herzog and Georgia Koppe
    A data‐driven closed‐loop control approach to drive neural state transitions for mechanistic insight
    Human Brain Mapping 47(11), 2026
    bioRxiv:2025.07.21.665992
    doi:10.1002/hbm.70600
    bibtex 
    @ARTICLE{EmondsHerbergGerchenPritschRochaZamoscikKirschHerzogKoppe:2026:1,
  AUTHOR = {Emonds, Niklas and Herberg, Evelyn and Gerchen, Martin Fungisai and Pritsch, Marc and Rocha, Joshua and Zamoscik, Vera and Kirsch, Peter and Herzog, Roland and Koppe, Georgia},
  PUBLISHER = {Wiley},
  DATE = {2026-07},
  DOI = {10.1002/hbm.70600},
  EPRINT = {2025.07.21.665992},
  EPRINTTYPE = {bioRxiv},
  JOURNALTITLE = {Human Brain Mapping},
  NUMBER = {11},
  TITLE = {A data‐driven closed‐loop control approach to drive neural state transitions for mechanistic insight},
  VOLUME = {47},
}
  • Leonie Kreis, Evelyn Herberg, Frederik Köhne, Anton Schiela and Roland Herzog
    SensLI: sensitivity-based layer insertion for residual and feedforward neural networks
    Journal of Machine Learning, 2026
    arXiv:2311.15995
    doi:10.4208/jml.251016
    bibtex 
    @ARTICLE{KreisHerbergKoehneSchielaHerzog:2026:1,
  AUTHOR = {Kreis, Leonie and Herberg, Evelyn and Köhne, Frederik and Schiela, Anton and Herzog, Roland},
  DATE = {2026-07},
  DOI = {10.4208/jml.251016},
  EPRINT = {2311.15995},
  EPRINTTYPE = {arXiv},
  JOURNALTITLE = {Journal of Machine Learning},
  TITLE = {SensLI: sensitivity-based layer insertion for residual and feedforward neural networks},
}

Latest Software

  • Lukas Baumgärtner, Roland Herzog, Stephan Schmidt and Manuel Weiß
    The proximal map of the weighted mean absolute error, 2023
    doi:10.5281/zenodo.7620815
    bibtex 
    @SOFTWARE{BaumgaertnerHerzogSchmidtWeiss:2022:2,
  AUTHOR = {Baumgärtner, Lukas and Herzog, Roland and Schmidt, Stephan and Weiß, Manuel},
  DATE = {2023},
  DOI = {10.5281/zenodo.7620815},
  TITLE = {The proximal map of the weighted mean absolute error},
}
  • Roland Herzog
    The SCOOP template engine, 2023
    https://pypi.org/project/scoop-template-engine/
    bibtex 
    @SOFTWARE{Herzog:2023:1,
  AUTHOR = {Herzog, Roland},
  URL = {https://pypi.org/project/scoop-template-engine/},
  DATE = {2023},
  TITLE = {The SCOOP template engine},
}
  • Robert Altmann and Roland Herzog
    Continuous Galerkin schemes for semi-explicit differential-algebraic equations, 2021
    doi:10.5281/zenodo.4682720
    bibtex 
    @SOFTWARE{AltmannHerzog:2021:1,
  AUTHOR = {Altmann, Robert and Herzog, Roland},
  DATE = {2021},
  DOI = {10.5281/zenodo.4682720},
  TITLE = {Continuous Galerkin schemes for semi-explicit differential-algebraic equations},
}
  • Tommy Etling, Roland Herzog, Estefanía Loayza and Gerd Wachsmuth
    First and second order shape optimization based on restricted mesh deformations, 2018
    doi:10.5281/zenodo.2547481
    bibtex 
    @SOFTWARE{EtlingHerzogLoayzaWachsmuth:2018:2,
  AUTHOR = {Etling, Tommy and Herzog, Roland and Loayza, Estefanía and Wachsmuth, Gerd},
  DATE = {2018},
  DOI = {10.5281/zenodo.2547481},
  TITLE = {First and second order shape optimization based on restricted mesh deformations},
}

Selected Talks

  • Roland Herzog
    Krylov subspace methods with a twist, 2025
    doi:10.5281/zenodo.15864810
    bibtex 
    @ONLINE{Herzog:2025:1,
  AUTHOR = {Herzog, Roland},
  DATE = {2025-06-25},
  DOI = {10.5281/zenodo.15864810},
  NOTE = {Plenary presentation given at the 30th Biennial Numerical Analysis Conference, University of Strathclyde},
  TITLE = {Krylov subspace methods with a twist},
}

Recent Teaching

Currently Supervising

  • M.Sc. Thesis of Laura Saiz:
    Newton Method for Manifold-Valued Image Denoising
    M.Sc. Mathematik, Heidelberg University
    Supervision: Roland Herzog, Georg Müller, Antonia Stavemann, Anton Schiela and Laura Weigl
  • M.Sc. Thesis of Jan Hendrik Agatz:
    A modern derivation of quasi-Newton formulas
    M.Sc. Mathematik, Heidelberg University
    Supervision: Roland Herzog and Georg Müller
  • M.Sc. Thesis of Mara-Eliana Popescu:
    Denoising Equations: Continuous and Discrete Diffusion Models for Symbolic Regression
    M.Sc. Data and Computer Science, Heidelberg University
    Supervision: Ullrich Köthe, Roland Herzog, Paul Sägert and Viktor Martinek
  • B.Sc. Thesis of Yannick Kopp:
    Integer Sum of Linear Ratios Optimization Problems
    B.Sc. Mathematik
    Supervision: Roland Herzog and Georg Müller

Recent Events Organized