BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20251122T195648EST-6348Sa9ePn@132.216.98.100 DTSTAMP:20251123T005648Z DESCRIPTION:Title: Particle Filters and Finite Ensemble Kalman Filters in L arge Dimensions: Theory\, Applied Practice\, and New Phenomena\n\nAbstract :\n\nThe filtering and predictive skill for turbulent signals is often lim ited by the lack of information about the true dynamics of the system and by our inability to resolve the assumed dynamics with sufficiently high re solution using the current computing power. The classical Kalman filter is no longer computationally feasible in such a high dimensional context. Th is problem can often be resolved by exploiting the underlying multiscale s tructure\, applying the full Kalman filtering procedures only to the large scale variables\, and estimating the small scale variables with proper st atistical strategies\, including multiplicative inflation\, representation model error in the observations\, and crude localization. A new error ana lysis framework for different reduced random Kalman filters is established \, The classical tools for Kalman filters can be used as a-priori performa nce criteria for the reduced filters. In applications\, these criteria gua rantee the reduced filters are robust\, and accurate for small noise syste ms. They also shed light on how to tune the reduced filters for stochastic turbulence. A new class of particle filters\, clustered particle filters\ , is also introduced for high-dimensional nonlinear systems. The clustered particle filter captures non-Gaussian features of the true signal which a re typical in complex nonlinear dynamical systems such as geophysical syst ems.\n DTSTART:20190516T211500Z DTEND:20190516T221500Z LOCATION:Room 1140\, CA\, Pav. André-Aisenstadt SUMMARY:Distinguished Lecture: Andrew J. Majda (Courant Institute\, NYU) URL:/mathstat/channels/event/distinguished-lecture-and rew-j-majda-courant-institute-nyu-297054 END:VEVENT END:VCALENDAR