| bibtype |
J -
Journal Article
|
| ARLID |
0313233 |
| utime |
20240111140707.5 |
| mtime |
20090325235959.9 |
| WOS |
000259407400003 |
| DOI |
10.1109/TSP.2008.928969 |
| title
(primary) (eng) |
Variational Bayesian Filtering |
| specification |
| page_count |
11 s. |
| media_type |
www |
|
| serial |
| ARLID |
cav_un_epca*0256727 |
| ISSN |
1053-587X |
| title
|
IEEE Transactions on Signal Processing |
| volume_id |
56 |
| volume |
10 (2008) |
| page_num |
5020-5030 |
|
| title
(cze) |
Variační Bayesovská Filtrace |
| keyword |
Bayesian filtering |
| keyword |
particle filtering |
| keyword |
Variational Bayes |
| author
(primary) |
| ARLID |
cav_un_auth*0101207 |
| name1 |
Šmídl |
| name2 |
Václav |
| institution |
UTIA-B |
| full_dept |
Department of Adaptive Systems |
| fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
| author
|
| ARLID |
cav_un_auth*0021112 |
| name1 |
Quinn |
| name2 |
A. |
| country |
IE |
|
| source |
|
| cas_special |
| project |
| project_id |
1M0572 |
| agency |
GA MŠk |
| ARLID |
cav_un_auth*0001814 |
|
| research |
CEZ:AV0Z10750506 |
| abstract
(eng) |
The use of the Variational Bayes (VB) approximation in Bayesian filtering is studied, both as a means to accelerate marginalized particle filtering, and as a deterministic local (one-step) approximation. The VB method of approximation and its variants are reviewed. These variants provide a range of algorithms that can be used in a principled trade-off between quality of approximation and computational cost. In combination with marginalized particle filtering, they generalize previously published work on variational filtering, and they extend currently available methods for speeding up stochastic approximations in Bayesian filtering. In particular, the free-form nature of the VB approximation allows optimal selection of moments which summarize the particles. The performance of the various VB filtering schemes is illustrated in the context of a Gaussian model with a nonlinear sub-state, and a hidden Markov model. |
| abstract
(cze) |
Práce se zabývá aplicaví metody Variační Bayes (VB) v oblasti Bayesovské filtrace. Základy metody VB a jejích variant jsou stručně shrnuty. Varianty této metody nabízejí škálu možností kompromisu mezi kvalitou aproximace a výpočetními náklady. Jednou z možných aplikací VB je zrychlení výpočtu marginalizovaného particle filtru tím, že se nagenerované částice (particly) nahradí jejich výběrovými momenty. Optimalizační procedura VB určí jaké momenty to budou. Porovnání výkonu nové metody s ostatními je provedeno na příkladu odhadování Gausovského stavového modelu s nelineární částí stavu a na příkladu Skrytého Markovského řetězce. |
| reportyear |
2009 |
| RIV |
BC |
| permalink |
http://hdl.handle.net/11104/0164114 |
| mrcbT16-f |
3.485 |
| mrcbT16-g |
0.272 |
| mrcbT16-h |
6.7 |
| mrcbT16-i |
0.04646 |
| mrcbT16-j |
1.1 |
| mrcbT16-k |
15979 |
| mrcbT16-l |
541 |
| mrcbT16-q |
151 |
| mrcbT16-s |
2.728 |
| mrcbT16-y |
27.68 |
| mrcbT16-x |
3.25 |
| arlyear |
2008 |
| mrcbU34 |
000259407400003 WOS |
| mrcbU56 |
pdf |
| mrcbU63 |
cav_un_epca*0256727 IEEE Transactions on Signal Processing 1053-587X 1941-0476 Roč. 56 č. 10 2008 5020 5030 |
|