This page contains software developed by the team (both present and former members) and data sets collected/used by the team. The software is either of “pedagogical nature” or software that was used to generate examples published in our papers. The pedagogical software is there in order to show how various algorithm works in order to let you start playing around with this algorithms as fast as possible. Each piece of software is accompanied with references to papers describing the mathematics underlying the code.
Software - pedagogical
Sequential Monte Carlo methods (particle filter/smoother)
Rao-Blackwellized particle filter (RBPF)
Bayesian Wiener system identification
Expectation Maximization (EM) for linear system identification
Expectation Maximization (EM) for nonlinear system identification
Particle Metropolis Hastings (PMH)
Particle Gibbs with ancestor sampling
Particle Metropolis Hastings using gradient and Hessian information
Software - used in papers
SMC for nonlinear system identification
SMC for graphical models
Rao-Blackwellized PSAEM
Robust particle filters via multiple importance sampling
Data sets