Machine Learning In Data Assimilation

Abstract:

Data assimilation refers to a particular class of inverse problems in which the unknown parameter is the initial condition, or entire solution trajectory, of a (possibly stochastic) dynamical system. The data comprises partial and noisy observations of the trajectory. Often, but not always, the goal of data assimilation is to perform the parameter inference sequentially as data is acquired. A primary use of data assimilation is in forecasting, where the purpose
is to provide better future estimates than can be obtained using either the data or the dynamical systems model alone. This lecture will overview
recent advances in the use of machine learning to enhance, or develop new, data assimilation algorithms.

Andrew Stuart - Bio

After postdoctoral research in applied mathematics at Oxford and MIT, Stuart held permanent positions at the University of Bath (1989–1992), in mathematics, at Stanford University (1991–1999), in engineering, and at Warwick University (1999–2016), in mathematics. He is currently Bren Professor of Computing and Mathematical Sciences at the California Institute of Technology. In 2009 he was elected an inaugural fellow of SIAM, and in 2020 he was elected a Fellow of the Royal Society. He was named a Vannevar Bush Faculty Fellow, by the Department of Defense, in 2022.