Seminars

Home Schools Computational Sciences Seminars

ABSTRACT: Reservoir computing has emerged as a machine learning architecture particularly suitable for modeling and predicting nonlinear and complex dynamical systems. In particular, as a recurrent neural network, it is itself a dynamical process and has been shown to serve as a digital twin of the studied dynamical system. The speaker will present the basics of reservoir computing and discuss several applications, including predicting chaotic ecosystems with limited data and anticipating tipping points in dynamical systems.