EMS TAG SciML 2026
Schedule
Tuesday, March 10 | Conference Room, Ground Floor, Seminar Building
| 12:00–12:30 | Registration |
| 12:30–12:45 | Welcome |
| 12:45–13:30 | Keynote: Elisa Riccietti Frequency-aware multigrid training in PINNs |
| 13:30–14:15 | Keynote: Janine Weber-Hamacher Machine learning-enhanced domain decomposition preconditioners ... and domain decomposition for numerically efficient machine learning |
| 14:15–14:45 | Coffee Break |
| 14:45–15:30 | Keynote: Silke Glas Machine learning enhanced structure-preserving model reduction: From the formulation on manifolds to data-driven realizations |
| 15:30–15:45 | Short Break |
| 15:45–16:10 | Contributed Talk: Stefano Pagani Geometry-aware Scientific Machine Learning: from Shape Reconstruction to Surrogate Models |
| 16:10–16:35 | Contributed Talk: Parvaneh Joharinad Geometric Evaluation of Representation of Data |
| 16:35–17:00 | Contributed Talk: Xenia Kobeleva Digital Twins in Neurology: Parameter inference and model selection for complex systems |
| 17:00–17:15 | Short Break |
| 17:15–17:40 | Contributed Talk: Vitalii Aksenov Accelerated Fixed-point Iteration over Spaces of Probability Measures |
| 17:40–18:05 | Contributed Talk: Konrad Janik Parametric SympNets |
| 18:05–18:30 | Contributed Talk: Jan Blechschmidt Flow-based Generative Modeling for Bayesian Inverse Problems |
| 18:30–21:00 | Welcome Reception |
Wednesday, March 11 | S21, Second Floor, Seminar Building
| 9:30–10:15 | Keynote: Paola Antonietti Machine Learning–Enhanced Polytopal Finite Element Methods with Applications to Neurodegenerative Disease Modelling |
| 10:15–11:00 | Keynote: Francesco Romor Non-parametric shape variability in SciML: applications to inter-patient hemodynamics |
| 11:00–11:30 | Coffee Break |
| 11:30–12:15 | Keynote: Carlo Marcati Approximation theory for neural and polynomial operator surrogates |
| 12:15–13:00 | Keynote: Kateryna Morozovska $PINN - domain decomposition with Bayesian Physics-Informed Neural Networks |
| 13:00–15:00 | Lunch & Poster Session Posters Poster – Theofanis Ifaistos: Generalization capabilities of Transformers in fluid dynamics Poster – Robin Pierschke: Deep Learning–Based Detection of HPV-Associated OPSCC from Histopathology Poster – Georg Winkler: Deep operator networks as surrogate models for computational fluid dynamics simulations Poster – Yujie Gong: A learning-enhanced nonlinear preconditioner for some hyperelasticity problems Poster – Carlotta Filippin: Graph-Based Nonlinear Reduced-Order Modeling for Time-Domain Electromagnetics Poster – David Oexle: A physics-encoded Fourier neural operator approach for surrogate modeling of divergence-free stress fields in solids Poster – Sophia Horak: Improving Numerical Schemes for Hyperbolic PDEs with Machine Learning Poster – Natalie Kubicki: A Flow-rate-conserving CNN-based Domain Decomposition Method for Blood Flow Simulation Poster – Lucas Mager: Mixture of Experts Surrogate Model for the Homogenization of Open-Porous Materials Poster – Umur Efe Arun: Physics-Constrained Hybrid Surrogate Models for Accelerated Parametric Exploration of Hyperelastic Structures |
| 15:00–15:25 | Contributed Talk: Gianluigi Rozza Surrogate modelling in parametric turbulent CFD: Model Reduction and Scientific Machine Learning |
| 15:25–15:50 | Contributed Talk: Julia Pelzer Resolving Extreme Data Scarcity by Explicit Physics Integration: An Application to Groundwater Heat Transport |
| 15:50–16:15 | Contributed Talk: Sebastian Götschel Hard-constraining Boundary Conditions for Physics-Informed Neural Operators |
| 16:15–16:45 | Coffee Break |
| 16:45–17:10 | Contributed Talk: Victor Michel-Dansac Neural semi-Lagrangian method for high-dimensional advection-diffusion problems |
| 17:10–17:35 | Contributed Talk: Giovanni Ziarelli An Hybrid Neural-Differential Framework for Parameter Discovery |
| 17:35 | Group Photo |
| 18:00 | Business Meeting |
Thursday, March 12 | S21, Second Floor, Seminar Building
| 9:30–10:15 | Keynote: Bruno Despres Autodiff of Neural networks and application to DeepRitz and FEM |
| 10:15–11:00 | Keynote: Samuel Leweke Challenges in Gray-box Chromatography Modeling |
| 11:00–11:30 | Coffee Break |
| 11:30–12:15 | Keynote: Paola Cinnella Towards high-fidelity quality aerodynamic optimization via multi-fidelity scientific machine learning |
| 12:15–13:00 | Keynote: Niccolò Tonicello Data-Driven Methods for Turbulence Modeling and Simulation |
| 13:00–13:15 | Closing |
| 13:15 | Farewell Lunch |
Contact
Email: cds-workshop(at)uni-koeln.de