Talk / Overview
We study the problem of rare event prediction for a class of slow-fast nonlinear dynamical systems. The state of the system of interest is described by a slow process, whereas a faster process drives its evolution. By taking advantage of recent advances in machine learning, we present a data-driven method to predict the future evolution of the state. We show that the method allows confident prediction of a rare transition event in a bi-stable system driven by a fast Lorenz-63 system at least several time steps in advance.