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. / CNDM / CADICS
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:liu-93461 |
| Date | January 2013 |
| Creators | Taghavi, Ehsan, Lindsten, Fredrik, Svensson, Lennart, Schön, Thomas B. |
| Publisher | Linköpings universitet, Reglerteknik, Linköpings universitet, Tekniska högskolan, Linköpings universitet, Reglerteknik, Linköpings universitet, Tekniska högskolan, School of Computational Science and Engineering, McMaster University, Division of Signals and Systems, Chalmers University |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Conference paper, info:eu-repo/semantics/conferenceObject, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | Proceedings of the 38th International Conference on Acoustics, Speech, and Signal Processing (ICASSP), p. 6293-6297 |
Page generated in 0.0024 seconds