Dr Thomas Schön
My main research interest is nonlinear inference problems, especially within the
context of dynamical systems, solved using probabilistic methods. In terms of scientific
fields, my research is situated somewhere on the intersection between the fields of
signal processing, machine learning, automatic control and mathematical statistics. My
aim is to pursue both basic and applied research, where the latter is typically carried
out in close collaboration with industry.
Recent Research Results/News |
- Jan 30, 2012: Together with Adrian Wills (University of Newcastle, Australia) and Bhushan
Gopaluni (University of British Columbia, Canada) I am organizing an invited
session on "Nonlinear System Identification and State Estimation using Sequential
Monte Carlo Methods" at the 16th IFAC Symposium on System Identification (SYSID) to be held
in Brussels, Belgium during July 11-13 this summer.
- Jan 30, 2012: We have some new results accepted for publication at the 16th IFAC Symposium on
System Identification (SYSID) to be held in Brussels, Belgium during July 11-13
this summer.
[1] Fredrik Lindsten, Thomas B. Schön and Michael I. Jordan,
A semiparametric Bayesian approach to Wiener system identification.
[2] Johan Dahlin, Fredrik Lindsten, Thomas B. Schön and Adrian Wills,
Robust ARX models with automatic order determination and Student's t innovations.
[3] Soren Henriksen, Adrian Wills, Thomas B. Schön and Brett Ninness,
Parallel Implementation of Particle MCMC Methods on a GPU.
[4] Fredrik Lindsten, Thomas B. Schön and Lennart Svensson,
A non-degenerate Rao-Blackwellised particle filter for estimating static parameters
in dynamical models.
[5] Adrian Wills, Thomas B. Schön, Fredrik Lindsten and Brett Ninness,
Estimation of Linear Systems using a Gibbs Sampler.
- Jan 24, 2012: Together with Moritz Diehl and the
OPTEC group I am organizing
a workshop on Moving Horizon Estimation and System Identification in Leuven, Belgium on
August 29-30, 2012. Some preliminary information about the workshop can be found here.
Click here here for more news.
|
