Staff Dr. Bastian Bohn

Address:
Institut für Numerische Simulation
Friedrich-Hirzebruch-Allee 7
53115 Bonn
Phone: +49 228 73-69813
Office: FHA7 3.005
E-Mail: ed tod nnob-inu tod sni ta nhoba tod b@foo tod de

Profile

  • Geschäftsführer (Akademischer Rat)

  • Budgetmanager

Thesis supervision

If you are interested in writing a Bachelor or Master thesis with me, please send me an e-mail. Possible topic areas include (among others)

  • mathematics of machine learning,

  • approximation theory,

  • high-dimensional discretizations.

News

Teaching

Winter semester 2025/26

Summer semester 2025

See teaching activities of the whole group.

Research Projects

Current

Multilevel sparse tensor product approximation for manifolds and for functions and operators on manifolds

Project C04, DFG SFB 1060.

Homepage.

Completed

SIMDATA-NL: Nichtlineare Charakterisierung und Analyse von FEM-Simulationsergebnissen für Autobauteile und Crash-Tests

BMBF support program.

Homepage.

Publications

Books

  1. Algorithmic Mathematics in Machine Learning. B. Bohn, J. Garcke, and M. Griebel. Data Science. SIAM, Philadelphia, PA, USA, 2024. BibTeX DOI

Journal papers

  1. Deep neural networks and PIDE discretizations. B. Bohn, M. Griebel, and D. Kannan. SIAM Journal on Mathematics of Data Science, 4(3):1145–1170, 2022. Also available as INS Preprint No. 2102. BibTeX Supplement PDF DOI
  2. Explainable Machine Learning for Scientific Insights and Discoveries. R. Roscher, B. Bohn, M. F. Duarte, and J. Garcke. IEEE Access, 8(1):42200–42216, 2020. BibTeX DOI arXiv
  3. A representer theorem for deep kernel learning. B. Bohn, M. Griebel, and C. Rieger. Journal of Machine Learning Research, 20(64):1–32, 2019. Also available as INS Preprint No. 1714. BibTeX PDF JMLR
  4. Error estimates for multivariate regression on discretized function spaces. B. Bohn and M. Griebel. SIAM Journal on Numerical Analysis, 55(4):1843–1866, 2017. Also available as INS Preprint No. 1412. BibTeX PDF DOI
  5. A sparse grid based method for generative dimensionality reduction of high-dimensional data. B. Bohn, J. Garcke, and M. Griebel. Journal of Computational Physics, 309:1–17, 2016. also available as INS Preprint No. 1514. BibTeX PDF DOI
  6. Feasible and successful: genome-wide interaction analysis (gwia) involving all \(1.9 \times 10^{11}\) pair-wise interaction tests. M. Steffens, T. Becker, T. Sander, R. Fimmers, C. Herold, D. Holler, C. Leu, S. Herms, S. Cichon, B. Bohn, T. Gerstner, M. Griebel, M. Nöthen, T. Wienker, and M. Baur. Human Heredity, 69:268–284, 2010. BibTeX Link

Refereed proceedings and book contributions

  1. Explain it to me – Facing Remote Sensing Challenges in the Bio- and Geosciences with Explainable Machine Learning. R. Roscher, B. Bohn, M. F. Duarte, and J. Garcke. In ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, volume V-3-2020, 817–824. 2020. BibTeX DOI
  2. Optimally rotated coordinate systems for adaptive least-squares regression on sparse grids. B. Bohn, M. Griebel, and J. Oettershagen. In Proceedings of the 2019 SIAM International Conference on Data Mining, 163–171. 2019. Also available as INS Preprint No. 1812. BibTeX PDF DOI
  3. On the convergence rate of sparse grid least squares regression. B. Bohn. In J. Garcke, D. Pflüger, C. Webster, and G. Zhang, editors, Sparse Grids and Applications - Miami 2016, volume 123 of Lecture Notes in Computational Science and Engineering, pages 19–41. Springer, 2018. BibTeX PDF
  4. Analysis of Car Crash Simulation Data with Nonlinear Machine Learning Methods. B. Bohn, J. Garcke, R. Iza-Teran, A. Paprotny, B. Peherstorfer, U. Schepsmeier, and C.-A. Thole. Procedia Computer Science, 18:621 – 630, 2013. Proceedings of the International Conference on Computational Science, ICCS 2013. BibTeX DOI
  5. An adaptive sparse grid approach for time series predictions. B. Bohn and M. Griebel. In J. Garcke and M. Griebel, editors, Sparse grids and Applications, volume 88 of Lecture Notes in Computational Science and Engineering, pages 1–30. Springer, 2012. BibTeX PDF DOI

Theses

  1. Error Analysis of regularized and unregularized Least-Squares Regression on discretized Function Spaces. B. Bohn. Dissertation, Institut für Numerische Simulation, Universität Bonn, 2017. BibTeX PDF
  2. Einbettung von Zeitreihen nach Takens' Theorem. B. Bohn. Diploma thesis, Institut für Numerische Simulation, Universität Bonn, 2010. BibTeX PDF

Technical reports and other non-refereed publications

  1. SIMDATA-NL - Nichtlineare Charakterisierung und Analyse von FEM-Simulationsergebnissen für Autobauteile und Crash-Tests. M. Griebel, H.-J. Bungartz, C. Czado, J. Garcke, U. Trottenberg, C.-A. Thole, B. Bohn, R. Iza-Teran, A. Paprotny, B. Peherstorfer, and U. Schepsmeier. Abschlussbericht des BMBF-Projekts, BMBF-Projekt, 2014. BibTeX PDF