Friday, February 26, 2021 1:30pm to 2:30pm
About this Event
The modeling and simulation of complex systems is critical for addressing problems ranging from drone aerodynamics to forecasting the spread of COVID-19. In the last decades our simulation capabilities have received a boost from unprecedented advances in computing hardware. However, simulations that can resolve all scales of these systems are impossible for the foreseeable future. But there is hope: these advances in computing enable novel learning algorithms that can assist human insight in the identification and control of complex systems. I will present our efforts to develop a computational framework to learn and control the effective dynamics of complex multi-scale systems. I will discuss its key algorithmic components, and their fusion, using examples in optimal testing for COVID-19, accelerated simulations of protein evolution and fish schooling.
0 people are interested in this event