skip to content

Center for Data and Simulation Science

Workshop Programme

Wednesday

12.00-13.00 Registration
13.00-13.15 Welcome
13.15-14.15 Jan S. Hesthaven Precision Algorithms: Algorithm Augmentation by Machine Learning
14.15-14.45 Marius Kurz Deep neural networks for data-driven LES closure models
14.45-15.15 Break
15.15-15.45 Simone Deparis Data driven approximation of parametrized PDEs by Reduced Basis and Neural Networks
15.45-16.15 Markus Pfeil Approximation of a marine ecosystem model using artificial neural networks
16.15-16.45 Break
16.45-17.15 Axel Klawonn Machine Learning in Adaptive Domain Decomposition Methods
17.15-17.35 Janine Weber Combining machine learning and adaptive coarse spaces to design robust and efficient FETI-DP methods in 3D
17.35-17.55 Francisco Meirinhos Exact renormalization group flow with neural-network parametrized effective action

Thursday

9.00-10.00 Eric Chung Learning multiscale models using nonlocal upscaling techniques
10.00-10.30 Nils Margenberg A deep learning based multigrid multiscale method for solving the Navier-Stokes-equations
10.30-11.00 Break
11.00-11.30 Toni Volkmer NFFT meets Krylov methods: Fast matrix-vector products for the graph Laplacian of fully connected networks
11.30-12.00 Rodrigo Iza Teran A Data Reduction Approach for Analysis of Simulation Bundles
12.00-12.20 Harish P. Cherukuri Artificial Neural Networks as Predictive Tools in Machining
12.20-13.45 Lunch Break
13.45-14.45 Xiao-Chuan Cai Modeling blood flow in patient-specific arteries
14.45-15.15 Alexander Heinlein Flow predictions using convolutional neural networks
15.15-15.45 Break
15.45-16.15 Luca Dedè An Artificial Neural Network based approach to Model Order Reduction of time-dependent models: application to multiscale cardiac simulation
16.15-16.45 Maximilian Merkert Model-Enhanced Machine Learning and Vice Versa
16.45-17.15 Break
17.15-17.45 Andrea Manzoni A comprehensive deep learning-based approach for reduced order modeling of nonlinear time-dependent parametrized PDEs
17.45-18.15 Martin Hess Reduced Order Models for Bifurcation Problems in Computational Fluid Mechanics with Autonomous Localization through Machine Learning
19.00 Conference Dinner at Brauhaus Päffgen

Friday

9.00-10.00 Eric C. Cyr Improved Neural Network Training: Layer-Parallelism, Least-squares and Initialization
10.00-10.30 Stefan Turek Machine Learning Technology and Techniques for Numerical Simulations with PDEs
10.30-11.00 Break
11.00-11.30 Patrick Buchfink Physics-preserving Projections for Compression of Data and Dynamical Systems
11.30-12.00 Jan Sokolowski On data driven system identification with artificial neural networks
12.00-12.30 Alexander Rüttgers Comparison of clustering techniques for hybrid rocket fuel combustion data
12.30-14.00 Farewell & Lunch