Rao-Blackwellised Particle Filtering for Dynamic Bayesian Networks
This work addresses the challenge of computational efficiency in inference for dynamic Bayesian networks, which is important for applications like robotics and regression, though it is incremental as it builds on existing particle filtering techniques.
The paper tackles the problem of improving the efficiency of particle filtering in dynamic Bayesian networks by using Rao-Blackwellisation, which combines sampling with exact marginalization, and shows that this method leads to more accurate estimates than standard particle filters.
Particle filters (PFs) are powerful sampling-based inference/learning algorithms for dynamic Bayesian networks (DBNs). They allow us to treat, in a principled way, any type of probability distribution, nonlinearity and non-stationarity. They have appeared in several fields under such names as "condensation", "sequential Monte Carlo" and "survival of the fittest". In this paper, we show how we can exploit the structure of the DBN to increase the efficiency of particle filtering, using a technique known as Rao-Blackwellisation. Essentially, this samples some of the variables, and marginalizes out the rest exactly, using the Kalman filter, HMM filter, junction tree algorithm, or any other finite dimensional optimal filter. We show that Rao-Blackwellised particle filters (RBPFs) lead to more accurate estimates than standard PFs. We demonstrate RBPFs on two problems, namely non-stationary online regression with radial basis function networks and robot localization and map building. We also discuss other potential application areas and provide references to some finite dimensional optimal filters.