The MOP - Particles without Resampling

Add to your list(s) Download to your calendar using vCal

• Maskell, S (University of Liverpool)
• Thursday 15 May 2014, 11:00-12:00
• Seminar Room 1, Newton Institute.

If you have a question about this talk, please contact Mustapha Amrani.

Advanced Monte Carlo Methods for Complex Inference Problems

Under certain assumptions, it is demonstrated that it is possible to derive a particle filter for arbitrarily high dimensions where the weights on the particles are all equal and there is therefore never any need for resampling. Relaxing these assumptions (eg by using ideas from “particle flow” implemented using piecewise linear approximations to a non-linear function) results in particle filters that only require very infrequent resampling. Results are demonstrated on non-linear non-Gaussian problems.

This talk is part of the Isaac Newton Institute Seminar Series series.

This talk is included in these lists:

• All CMS events
• Featured lists
• INI info aggregator
• Isaac Newton Institute Seminar Series
• School of Physical Sciences
• Seminar Room 1, Newton Institute
• bld31

Note that ex-directory lists are not shown.