In this talk, I will present recent research that builds fast and accurate data-driven surrogate models (or emulators) for various complex and high-dimensional systems. Furthermore we will use scientific machine learning techniques in lieu of black-box data-driven methods. In other words, not only will our models be informed by data, but they will also be informed by inductive biases that are designed to preserve some a priori knowledge of the underlying structure of the problem (for example inertial or conservative dynamics, large Kolmogorov width, chaotic or multiscale behavior). We will discuss examples of canonical systems where scientific machine learning provides superior outcomes when compared to classical surrogate modeling strategies. Furthermore, in addition to canonical experiments, our algorithms will be demonstrated for several real-world applications of strategic importance, for example for data-driven geophysical forecasting from ship and satellite observation data and wind-turbine wake predictions from meteorological and LIDAR measurements.

Bio: