Adaptive Stopping for Fast Particle Smoothing
Paper in proceeding, 2013

Particle smoothing is useful for offline state inference and parameter learning in nonlinear/non-Gaussian state-space models. However, many particle smoothers, such as the popular forward filter/backward simulator (FFBS), are plagued by a quadratic computational complexity in the number of particles. One approach to tackle this issue is to use rejection-sampling-based FFBS (RS-FFBS), which asymptotically reaches linear complexity. In practice, however, the constants can be quite large and the actual gain in computational time limited. In this contribution, we develop a hybrid method, governed by an adaptive stopping rule, in order to exploit the benefits, but avoid the drawbacks, of RS-FFBS. The resulting particle smoother is shown in a simulation study to be considerably more computationally efficient than both FFBS and RS-FFBS.

backward simulation

CHAIN MONTE-CARLO

Sequential Monte Carlo

particle smoothing

Author

Ehsan Taghavi

Chalmers, Signals and Systems

F. Lindsten

Linköping University

Lennart Svensson

Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering

T. B. Schon

Linköping University

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

15206149 (ISSN)

6293-6297
978-1-4799-0356-6 (ISBN)

Subject Categories

Electrical Engineering, Electronic Engineering, Information Engineering

DOI

10.1109/ICASSP.2013.6638876

ISBN

978-1-4799-0356-6

More information

Latest update

10/5/2023