An exploration of data-driven techniques for predicting extreme events in intermittent dynamical systems

Abstract

The ability to characterize and predict extreme events is a vital topic in fields ranging from finance to ocean engineering. Typically, the most-extreme events are also the most-rare, and it is this property that makes data collection and direct simulation challenging. In this thesis, I will develop a data-driven objective, alpha-star, appropriate for optimizing extreme event predictor schemes. This objective is constructed from the same principles as Reciever Operating Characteristic Curves, and exhibits a geometric connection to scale separation. Additionally, I will demonstrate the application of alpha-star to the advance prediction of intermittent extreme events in the Majda-McLaughlin-Tabak model of a dispersive fluid.