Automated model building and representation learning for multiscale simulations (B07)

• Subject Area: Statistical Physics, Nonlinear Dynamics, Complex Systems, Soft and Fluid Matter, Biological Physics
• Term: since 2018
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 233630050

The project addresses applications of machine learning techniques to multi-scale simulation of soft-matter systems. We focus on two aspects: The first is to build a coarser-grained model, e.g., using generic function approximators such as kernels or deep neural networks to learn latent representations of the coarser level dynamics from data. The second aspect is back-mapping, i.e., the reconstruction of a high-resolution simulation state corresponding to the coarse-grained state. In the next funding period, one main goal for both aspects will be efficiency. In addition, we will explore new applications for backmapping, where the coarse-scale information might come from different data sources.

• DFG Programme: CRC/Transregios
• Subproject of: TRR 146: Multiscale Simulation Methods in Soft Matter Systems
• International Connection: Netherlands
• Applicant Institution: Johannes Gutenberg-Universität Mainz
• Co-Applicant Institution: Max-Planck-Institut für Polymerforschung
• Project Heads: Dr. Denis Andrienko, since 7/2022; Professor Dr. Tristan Bereau, until 6/2022; Professor Dr. Michael Wand