*arrow_back*Back

## Propagation of Uncertainty and Chaos in Multiscale Systems

#### Mainz

#### Organisers

**PLEASE NOTE: DUE TO THE RISING COVID NUMBERS IN GERMANY, THE EVENT WILL BE A PURE ONLINE EVENT.**

**WE WILL UPDATE PARTICIPANTS IF THIS CHANGES.**

The purpose of this workshop is to bring together experts on different areas such as numerics, computer science and statistical physics to gain a deeper insight into the different analytical and numerical methods for multiscale problems. Modeling and simulation of multiscale systems constitutes a grand challenge in computational science, and is widely applied in fields ranging from the physical sciences and engineering to life science. In physics, multiscale modeling is aimed at calculating material properties or the dynamics of systems on one level, using information or models at different levels. Here the levels mostly are descriptions of a the system at certain scales or resolutions, e.g. atomic, mirco, macro scales. If one has descriptions of different levels, for instance, micro or macro level, the aim is to use the information from lower levels in the higher level description without having to resolve the lower level completely. Due to the random nature of many mircolevels one has to take the uncertainty of the data and the description into account. This can range from simple equations with random coefficients to stochastic differential equations. Since the main objective of the whole TRR is coarse-graining we want to gain a deeper knowledge of one possible method for this, namely multiscale modeling. We want to focus on the following topics:

• How to construct and solve multiscale systems (Methods, Approximations)?

• Can we reduce the equations, if we are only interested in the expected value of the solutions (Closure, Kinetic Theory)?

• How does mathematical statistical mechanics come into play (Gibbs measures, Gibbsian point processes, Multi body dynamics) ?

• How does one apply Data-Driven Methods in multiscale problems with respect to physical properties (Machine Learning, Deep Learning, Tensor trees)?

## References

**Germany**

Aaron Brunk (Johannes Gutenberg University) - Organiser

Fabio Frommer (Johannes Gutenberg-Universität) - Organiser