Lookahead in Sequential Monte Carlo

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

• Prof Rong Chen, Department of Statistics, Rutgers University
• Wednesday 25 November 2009, 14:30-15:15
• LR4, Engineering, Department of.

If you have a question about this talk, please contact Rachel Fogg.

Dynamic systems and state space models are widely used in many fields of applications. Most of them are nonlinear and non-Gaussian. Sequential Carlo methods (Monte Carlo filters) have been developed for dealing with such systems efficiently. It is noticed that most of the dynamic systems process strong memory. In such systems, future observations often contain significant amount of information on the current system state. Hence, lookahead (estimation with future observations) can often produce more accurate inferences on the current state. In this talk we present several efficient lookahead Sequential Monte Carlo algorithms that utilize the future information with manageable computational cost. Some theoretical justifications and applications in signal processing will be presented.

This talk is part of the Probabilistic Systems, Information, and Inference Group Seminars series.

This talk is included in these lists:

• All Talks (aka the CURE list)
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge University Engineering Department Talks
• Cambridge talks
• Centre for Smart Infrastructure & Construction
• Chris Davis' list
• Computational Continuum Mechanics Group Seminars
• Featured lists
• Information Engineering Division seminar list
• Interested Talks
• LR4, Engineering, Department of
• Probabilistic Systems, Information, and Inference Group Seminars
• School of Technology
• Trust & Technology Initiative - interesting events
• bld31
• ndk22's list
• ob366-ai4er
• rp587
• yk449

Note that ex-directory lists are not shown.