Florian Rossmannek
Assistant Professor | NTU Singapore
Open positions available, see here
About Me
I am a Nanyang Assistant Professor at NTU Singapore in the Division of Mathematical Sciences doing research on the mathematics of machine learning.
Prior to that, I was a Postdoctoral Fellow at NTU Singapore, mentored by Juan-Pablo Ortega.
I obtained my doctoral degree from ETH Zurich in March 2023, advised by Patrick Cheridito and Arnulf Jentzen.
I am an alumnus of the Schmidt AI in Science program and of the RiskLab Switzerland.
Recent Highlights
- 'Echoes of the past: A unified perspective on fading memory and echo states' (with J-P. Ortega) [arXiv] got accepted for publication in Neural Computation
- Invited talk at the NTU-HUN-REN SZTAKI machine learning workshop (picture)
- Research visit at the HUN-REN Institute for Computer Science and Control (SZTAKI), Budapest
- Joint workshop with NTU's EMPOWER Centre (program)
- 'Stochastic dynamics learning with state-space systems' (with J-P. Ortega) [arXiv] got accepted for publication in M3AS
Previous Highlights
-
2025
- Invited talk at the SPMS MAS Seminar Series, NTU
- Invited talk at the Differential Equations for Data Science online seminar
- Research visit at ETH Zurich
- Invited talk at the Quantitative Finance Conference 2025 @NUS (picture)
- Research visit at the University of Tokyo
- Start of the highlights section (July)
Open Positions
I am looking for Postdocs, PhD students, and Research Assistants.
Inquiries can be made anytime by email.
Interested applicants should have a strong mathematical background and be broadly interested in the following topics:
• dynamical systems • theory of machine learning • reservoir computing • neural networks
Research Interest L•DAWN
Learning • Dynamics Attractors With Networks
I work on the mathematical foundation of machine learning, where I am particularly interested in the mechanisms of models (such as reservoir computers, state-space systems, kernel methods, etc.) used for learning and forecasting of (deterministic and stochastic) dynamical systems.
My earlier work concerned approximation and optimization problems in (static) neural network theory.
Research Articles
Publications
- Fading memory and the convolution theorem (with J-P. Ortega)
IEEE Trans. Autom. Control, vol 70(12), 2025 [journal / arXiv]
- State-space systems as dynamic generative models (with J-P. Ortega)
Proc. R. Soc. A, vol 481(2309), 2025 [journal / arXiv]
- Gradient descent provably escapes saddle points in the training of shallow ReLU networks (with P. Cheridito and A. Jentzen)
J. Optim. Theory Appl., vol 203(3), 2024 [journal / arXiv]
- Landscape analysis for shallow neural networks: complete classification of critical points for affine target functions (with P. Cheridito and A. Jentzen)
J. Nonlinear Sci., vol 32(5), 2022 [journal / arXiv]
- A proof of convergence for gradient descent in the training of artificial neural networks for constant target functions (with P. Cheridito, A. Jentzen, and A. Riekert)
J. Complexity, vol 72, 2022 [journal / arXiv]
- Non-convergence of stochastic gradient descent in the training of deep neural networks (with P. Cheridito and A. Jentzen)
J. Complexity, vol 64, 2021 [journal / arXiv]
- Efficient approximation of high-dimensional functions with neural networks (with P. Cheridito and A. Jentzen)
IEEE Trans. Neural Netw. Learn. Syst., vol 33(7), 2022 [journal / arXiv]
Preprints
- Echoes of the past: A unified perspective on fading memory and echo states (with J-P. Ortega), accepted for publication in Neural Comp., 2025 [arXiv]
- Stochastic dynamics learning with state-space systems (with J-P. Ortega), accepted for publication in Math. Models Methods Appl. Sci., 2025 [arXiv]
- Efficient Sobolev approximation of linear parabolic PDEs in high dimensions (with P. Cheridito), 2023 [arXiv]
Miscellaneous
- PhD thesis: The curse of dimensionality and gradient-based training of neural networks: shrinking the gap between theory and applications (2023) [link]
- An exercise in combinatorics: Christmas Stars (2021) [pdf]
- MSc thesis: Magnetic and Exotic Anosov Hamiltonian Structures (2019)
Teaching
tba
Past Courses at ETH Zurich
| 2022 | coordinator, Mathematics I (401-0291-00L) |
| 2022 | coordinator, Probability and Statistics (401-2604-00L) |
| 2021 | coordinator, Mathematics II (401-0292-00L) |
| 2019 | coordinator, Seminar on the Theory and Applications of Machine Learning (D-MATH) |
| 2018 | TA, Analysis I (401-1261-07L) |
| 2017 | TA, Algorithms and Complexity (252-0851-00L) |
| 2017 | TA, Topology (401-2554-00L) |
| 2016 | TA, Algorithms and Complexity (252-0851-00L) |
Resume
Employment
| Since 2025 | Assistant ProfessorNTU Singapore |
| 2023 - 2025 | Postdoctoral FellowNTU Singapore |
| 2019 - 2023 | Research AssistantETH Zurich |
Education
| 2019 - 2023 | PhD MathematicsETH Zurich |
| 2017 - 2019 | MSc MathematicsETH Zurich |
| 2014 - 2018 | BSc MathematicsETH Zurich |
Fellowships / Grants
| 2025 | Nanyang Assistant Professorship Startup Research Grant |
| 2023 | Eric and Wendy Schmidt AI in Science Postdoctoral Fellowship |
Selected Talks
