Michael Graham, professor at the University of Wisconsin-Madison, delivered the 2023 Stephen H. Davis Lecture.

Graham's lecture lecture outlined how exploitation of symmetries substantially improves performance in accurately simulating chaotic or turbulent fluid flows. Graham described a data-driven reduced order modeling method, “Data-driven Manifold Dynamics,” that finds a nonlinear coordinate representation of the manifold using a machine-learning architecture called an autoencoder, then learns an ordinary differential equation for the dynamics on the manifold.