Prof. Dr. Felix Dietrich
Academic Career and Research Areas
Prof. Dietrich (b. 1989) conducts research in the analysis and development of numerical algorithms for machine learning. This includes algorithms to enable, accelerate, and optimize simulation and analysis of complex dynamical systems, nonlinear manifold learning techniques, and the connection of neural networks with Gaussian processes. His group focuses on kernel methods and data-driven approximations of Koopman and Laplace operators.
Prof. Dietrich studied Scientific Computing (B.Sc.) at UAS Munich and KTH in Stockholm. He obtained his Masters (2014) and PhD (2017) in Mathematics (TopMath) from the Technical University of Munich. From 2017 to 2019, he worked as a postdoc at JHU and at Princeton University, together with Prof. Kevrekidis. In 2019, he joined TUM to work at the chair of Scientific Computing, and starting 2022, as leader of an DFG Emmy Noether Junior Research Group. In 2024, he was appointed to the professorship for Physics-Enhanced Machine Learning at TUM.
Awards
- Emmy Noether Junior Research Group of the German Research Foundation (2021)
- Scholarship of the German Academic Scholarship Foundation (2009-2014)
- Scholarship of the Max Weber-Program of the state of Bavaria (2009-2014)
Key Publications (all publications)
E. Bolager, I. Burak, C. Datar, Q. Sun, F. Dietrich. "Sampling weights of deep neural networks". In: Advances in Neural Information Processing Systems 36 (NeurIPS 2023).
AbstractE. Peterfreund, O. Lindenbaum, F. Dietrich, T. Bertalan, M. Gavish, I. G. Kevrekidis, and R. R. Coifman. "Local conformal autoencoder for standardized data coordinates". In: Proceedings of the National Academy of Sciences (Nov. 2020), p. 202014627.
AbstractF. Dietrich, T. N. Thiem, and I. G. Kevrekidis. "On the Koopman Operator of Algorithms". In: SIAM Journal on Applied Dynamical Systems 19.2 (Jan. 2020), pp. 860–885.
AbstractT. Bertalan, F. Dietrich, I. Mezic, and I. G. Kevrekidis. "On learning Hamiltonian systems from data". In: Chaos 29.12 (Dec. 2019), p. 121107.
AbstractQ. Li, F. Dietrich, E. M. Bollt, and I. G. Kevrekidis. "Extended dynamic mode decomposition with dictionary learning: A data-driven adaptive spectral decomposition of the Koopman operator". In: Chaos 27.10 (Oct. 2017), p. 103111.
Abstract