Geometric and
structure-aware learning
We design graph neural networks and equivariant representations that incorporate geometric and relational structure, and analyze their properties.
- Graph neural networks
- Equivariance
University of Stuttgart · MLS Lab
Machine Learning and Simulation Lab
We are a research group at the University of Stuttgart. We develop and analyze new machine learning methods, with applications in the natural sciences and engineering.
Machine Learning and Simulation LabStuttgart, Germany
01 / Research
Research areas
Our work focuses on geometric and structure-aware learning, generative and probabilistic models, and machine learning for science and engineering.
Geometric and
Generative &
probabilistic models
We develop generative models and inference methods that incorporate structure and constraints, aiming for efficient sampling and controllable generation.
- Diffusion models
- Controllable generation
Machine learning for
science and engineering
We develop new physics-aware models and neural solvers for scientific simulation, with applications in computational chemistry and fluid dynamics.
- Neural PDE solvers
- Molecular learning
02 / Publications
Selected publications
SymDrift: One-Shot Generative Modeling under Symmetries
Samir Darouich, Vinh Tong, Lluís Pastor-Pérez, Tanja Bien, Loay Mualem, Mathias Niepert
SMART: Scalable Mesh-free Aerodynamic Simulations from Raw Geometries using a Transformer-based Surrogate Model
Jan Hagnberger, Mathias Niepert
GraphBench: Next-generation graph learning benchmarking
Timo Stoll, Chendi Qian, Ben Finkelshtein, Ali Parviz, Darius Weber, Fabrizio Frasca, Hadar Shavit, Antoine Siraudin, Arman Mielke, Marie Anastacio, Erik Müller, Maya Bechler-Speicher, Michael Bronstein, Mikhail Galkin, Holger Hoos, Mathias Niepert, Bryan Perozzi, Jan Tönshoff, Christopher Morris
Learning (Approximately) Equivariant Networks via Constrained Optimization
Andrei Manolache, Luiz Chamon, Mathias Niepert
CALM-PDE: Continuous and Adaptive Convolutions for Latent Space Modeling of Time-dependent PDEs
Jan Hagnberger, Daniel Musekamp, Mathias Niepert
See the publication archive for bibliographic details and available paper links.
03 / People
Group members
Group leader
Mathias Niepert
Professor · University of Stuttgart
Prof. Mathias Niepert leads the MLS Lab at the Institute for Artificial Intelligence. His research focuses on machine learning methods that incorporate physical, logical, and geometric structure.
He is an ELLIS Fellow, a faculty member of IMPRS-IS, and Chief Scientific Advisor at NEC Laboratories Europe.
Administrative staff
Research staff
Alumni
04 / Teaching & opportunities
