Jonas Latz, M.Sc.

Name : Jonas Latz Address : M2 - Lehrstuhl für Numerische Mathematik Boltzmannstraße 3 85748 Garching b.M. Office : 03.10.055 Phone : +49 89 289 18404 E-mail : jonas.latztum.de (Send me an encrypted email)

Research

Interests

Methods in Numerical Analysis², Computational Science, Probability Theory³ and Statistics¹ that can be used to blend mathematical models with observational data.

Publications

1. Christian Kahle, Kei Fong Lam, Jonas Latz, Elisabeth Ullmann (2018)
   Bayesian parameter identification in Cahn-Hilliard models for biological growth
   Under review. (Abstract, *.bib, arXiv )
2. Jonas Latz, Marvin Eisenberger, Elisabeth Ullmann (2018)
   Fast Sampling of parameterised Gaussian random fields
   Under review. (Abstract, *.bib, arXiv )
3. Jonas Latz, Iason Papaioannou, Elisabeth Ullmann (2018)
   Multilevel Sequential² Monte Carlo for Bayesian Inverse Problems
   J. Comput. Phys. 368, p. 154-178, DOI . (Abstract, *.bib, arXiv )

Talks

• (July 1st-6th 2018) 13th International Conference in Monte Carlo & Quasi-Monte Carlo Methods in Scientific Computing (MCQMC 2018) , Rennes, France.
• (April 16th-19th 2018) SIAM Conference on Uncertainty Quantification 2018 (UQ18) , Garden Grove, CA, USA.
• (March 19th-23rd 2018) GAMM Annual Meeting 2018 , Munich, Germany.
• (February 20th 2018) Seminar , Isaac Newton Institute, Cambridge, United Kingdom.
• (February 5th-9th 2018) Surrogate models for UQ in complex systems , Isaac Newton Institute, Cambridge, United Kingdom. (Poster)
• (November 22nd 2017) Seminar Stochastics, Statistics and Numerical Analysis, Universität Mannheim, Germany. (Slides)
• (September 6th-8th 2017) FrontUQ in Engineering, Munich, Germany.
• (June 15th-17th 2017) 2nd International Conference on Uncertainty Quantification in Computational Sciences and Engineering (UNCECOMP 2017) , Rhodes, Greece.
• (March 24th 2017) Group Seminar, Hybrid Materials Interfaces Group, Universität Bremen, Germany.

Stays

• (September - November 2018) Chair of Scientific Computing and Uncertainty Quantification, École Polytechnique Fédérale de Lausanne (EPFL), Switzerland. (with Fabio Nobile)
• (February 5th-23rd 2018) Isaac Newton Institute, University of Cambridge, UK.
• (November 20th-22nd 2017) Chair of Mathematical Optimization, Universität Mannheim, Germany. (with Claudia Schillings)

Teaching

Lectures

• (July 18th 2017) Guest lecture "Bayesian Inverse Problems" in Algorithms for Uncertainty Quantification, Chair for Scientific Computing, Computer Science, TUM.

Tutorials

• (Summer term 2018) Tutorials in Numerical Methods for Uncertainty Quantification,
• (Winter term 2017/18) Tutorials in Einführung in die Numerische Lineare Algebra,
• (Summer term 2017) Tutorials in Numerical Methods for Uncertainty Quantification and Supervision in Hauptseminar: Uncertainty Quantification with Efficient Monte Carlo Methods,
• (Winter term 2016/17) Tutorials in Maß- und Integrationstheorie,
• (Term 1/2 2015/16) Undergraduate Supervisor (Mathematics Institute, University of Warwick),
• (Summer term 2015) Tutorials in Elemente der Analysis 2 (Fachbereich IV, Universität Trier),
• (Winter term 2014/15) Tutorials in Elemente der Analysis 1 and Tutorials in Mikroökonomik (Fachbereich IV, Universität Trier),
• (Winter term 2013/14) Tutorials in Mikroökonomik (Fachbereich IV, Universität Trier).

Supervised students

• (ongoing) Sarah Kreuzer (B.Sc. student, working title: Efficient implementation of the Ising-model and image denoising)
• (ongoing) Sabrina Balzer (M.Sc. student, working title: Surrogate construction for a Cahn-Hillard model and Bayesian Inversion; with Christian Kahle and Elisabeth Ullmann)
• (ongoing) Fabian Wagner (M.Sc. student, working title: Particle Methods for Deep Learning)
• (2018) Matthieu Bulte (B.Sc. student, title: Sequential Monte Carlo for time-dependent Bayesian Inverse Problems)
• (2017) Marvin Eisenberger (M.Sc. student, title: Parameterized solution of Karhunen-Loève Eigenproblems with Reduced Basis Methods in Uncertainty Quantification, now: TUM Dep. Informatics)

Education

Degree programmes

• (ongoing) Doctoral Studies in Mathematics, Technische Universität München, Germany
• (2015-2016) Master of Science in Scientific Computing (with Distinction), University of Warwick, UK
• (2014-2016) Master of Science in Applied Mathematics and Economics, Universität Trier, Germany
• (2012-2014) Bachelor of Science in Business Mathematics, Universität Trier

Theses

• Jonas Latz (2016)
  Bayes Linear Methods for Inverse Problems
  Master's Thesis, University of Warwick. (*.pdf | *.bib)
• Jonas Latz (2014)
  Äußere Hausdorff Maße: Anwendungen und Eigenschaften
  Bachelor's Thesis, Universität Trier.

Useful(?!) Links

• Reading Group Uncertainty Quantification
• Project 10.02/BAYES@International Graduate School of Science and Engineering at TUM.
• Networks: GitHub\latz-io , orcid.org , researchgate 
• Lunch menus@Garching: FMI Cafeteria, IPP Canteen , Mensa 
• Science Communication:
  • Uncertaintea Quantification (Youtube) , Science Slam, Hamburg, 2017 (in German)
  • Latz, Ullmann: Bayes'sche Inversion am Beispiel des Fadenpendels, in Quartl 84, Garching, 2017 (in German)
  • Latz, Ullmann: Wie lernen Computer die Erdbeschleunigung?, in TUM Campus Magazin, 01/2018 (in German)