Bayesian inference for dynamic models with dirichlet process mixtures

Conference item

Using Kalman techniques, it is possible to perform optimal estimation in linear Gaussian state-space models. We address here the case where the noise probability density functions are of unknown functional form. A flexible Bayesian nonparametric noise model based on mixture of Dirichlet processes is introduced. Efficient Markov chain Monte Carlo and Sequential Monte Carlo methods are then developed to perform optimal estimation in such contexts.

Authors: Caron, F; Davy, M; Doucet, A; Duflos, E; Vanheeghe, P

• Publication status: Published
• Host title: 2006 9th International Conference on Information Fusion, Vols 1-4
• Pages: 138-145
• Publication date: 2006-01-01
• DOI: 10.1109/ICIF.2006.301580
• ISBN: 9781424409532
• Keywords: Bayesian nonparametrics; monte carlo markov chain; particle filter; dirichlet process mixture; rao-blackwellisation