![[Search]](/static/images/search.gif){kind=small}
![[A-Z Index]](/static/images/az.gif){kind=small}
![[Contact]](/static/images/contact.gif){kind=small}
![[University of Cambridge]](/static/images/identifier2.gif){kind=small}
![[Talks.cam]](/static/images/talkslogosmall.gif)
Peter Jan van Leeuwen, Prof in Data Assimilation, Department of Meteorology, University of Reading
Wednesday 25 January 2012, 14:15-15:00
Optimal proposal densities for particle filters
- đ¤ Speaker: Peter Jan van Leeuwen, Prof in Data Assimilation, Department of Meteorology, University of Reading
- đ Date & Time: Wednesday 25 January 2012, 14:15 - 15:00
- đ Venue: LR4, Engineering, Department of
Abstract
In this talk, we discuss why the so-called ‘optimal proposal density’in particle filtering is not efficient when the number of independent observations is large. Two new methods for constructing proposal
densities are introduced that ensure that all or most of the particles end up in the high probability regions with similar weights. These similar weights opens up the field of efficient particle filtering in
high-dimensional systems. Applications to simple systems and a high-dimensional highly nonlinear fluid-dynamics problem (the barotropic vorticity equation) will be discussed.
Series This talk is part of the Probabilistic Systems, Information, and Inference Group Seminars series.
Included in Lists
This talk is not included in any other list.
Note: Ex-directory lists are not shown.