Jornal article

Šmídl V., Quinn A.

IEEE Transactions on Signal Processing

2008

10 (2008)

5020-5030

1053-587X

Ing. Václav Šmídl, Ph.D. (ÚTIA - Oddělení adaptivních systémů)

Bayesian filtering, particle filtering, Variational Bayes

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.