Maximilian Dinkel, M.Sc.

Contact

• Room 1228
• Email: maximilian.dinkel@tum.de
• Phone: +49 (0) 89 289 15240
• Fax: +49 (0) 89 289 15301

Teaching

• Numerische Methoden für Ingenieure (Übung), WS 24/25

Research Interests

• Bayesian Methods
• Inverse Problems
• Gaussian Processes
• Physics Informed Machine Learning

Publications

• Maximilian Dinkel, Carolin M. Geitner, Gil Robalo Rei, Jonas Nitzler, Wolfgang A. Wall: Solving Bayesian Inverse Problems With Expensive Likelihoods Using Constrained Gaussian Processes and Active Learning, (2024), Inverse Problems, Volume 40.9, 095008, DOI 

International Conference Contributions with Abstract

• Maximilian Dinkel, Carolin M. Geitner, Wolfgang A. Wall: An approach for solving inverse problems using constrained Gaussian processes, UNCECOMP 2023, 11.-14.06.2023, Athens, Greece
• Maximilian Dinkel, Gil Robalo Rei, Wolfgang A. Wall: Dynamic Learning Rate Decay for Stochastic Variational Inference, 4th IMA Conference on Inverse Problems from Theory to Application, 11.-13.09.2024, Bath, UK

Supervised student projects / theses

• D. Still: Adjoint based Bayesian Inference of Random Fields with Hamiltonian Monte Carlo Samplers, Master's Thesis, 2023 (supervised together with Gil Robalo Rei)

Education

• Since 2022 Research Associate at the Institute for Computational Mechanics (Lehrstuhl für Numerische Mechanik), Technische Universität München, Germany
• 2022 Master of Science (M.Sc.), Mechanical Engineering, Technische Universität München, Germany
• 2019 Bachelor of Science (B.Sc.), Management and Technology, Technische Universität München, Germany
• 2018 Bachelor of Science (B.Sc.), Mechanical Engineering, Technische Universität München, Germany