ML for complex dynamical systems

For the description of complex dynamical systems, data-driven modeling and AI are gaining increasing importance. In this context, large data sets from experiments and computer simulations are processed and analyzed to build effective deterministic (differential equation) models, predict critical behavioral changes, and identify significant events.

Course

In the courses offered here, AI methods in the analysis of complex dynamical systems are conveyed in a comprehensible manner and applied to concrete modeling problems. Examples include the ML-based extraction of macroscopic evolution equations from molecular dynamics simulations of microscopic physical processes,  the anticipation of critical transitions in the dynamics of complex systems, or the control of nonlinear dynamical systems.

• Foundations of deterministic and stochastic dynamic systems

• Physics informed deep learning

• Inferring dynamical systems from data

• Control of dynamical systems