Skip to main content
Department of Information Technology

Sequential Monte Carlo methods

Description

The aim of this course is to provide an introduction to the theory and application of computational methods for inference in nonlinear dynamical systems, as well as more general problems. More specifically we will introduce sequential Monte Carlo (SMC) methods and Markov chain Monte Carlo (MCMC) methods and show how these can be used to solve challenging learning (system identification) and state estimation problems in nonlinear dynamical systems. We will also discuss SMC in a more general context, showing how it can be used as a generic tool for sampling from complex probability distributions.

This is an intensive course held during August 24-29, 2017 in conjunction to the SMC workshop which takes place during the time August 30 - September 1, 2017.

SMC_course_group_edit_small.JPG

Content

  • Probabilistic modeling of dynamical systems
  • Filtering and smoothing
  • The Monte Carlo idea and importance sampling
  • Particle filtering / Sequential Monte Carlo
  • Basic convergence theory for particle filters
  • Likelihood estimation and maximum likelihood parameter inference
  • Particle Markov Chain Monte Carlo
  • High-dimensional filtering
  • Generic SMC / SMC Samplers
  • SMC for probabilistic programming

The course (including successful completion of the homework assignments) corresponds to 6 ECTS credits/högskolepoäng.

Course Structure

The course consists of lectures and homework assignments. The homeworks will to a large extent be computer based, please bring your own laptop with Matlab installed (or another programming environment of your choice, e.g. Python, Julia, R...)

Examination

Via successfully completing and handing in the hand-in assignments.

Course literature

Lecture notes will be made available to the course participants,
Thomas B. Schön and Fredrik Lindsten. Learning of dynamical systems - Particle filters and Markov chain methods, Lecture notes, 2017. Available here.

Periodicity

Every 2 years. Earlier editions of this course have been given at ICASSP (2016), Universidad Tecnica Federico Santa Maria, Valparaiso, Chile (2014), KTH (2012), University of Sydney, Sydney, Australia (2012) and Vrije Universiteit Brussel, Brussels, Belgium, (2012).

Schedule

The course takes place on August 24-25 (Thursday-Friday) and August 28-29 (Monday-Tuesday) 2017, with a social event on Saturday, August 26.

See the schedule page for more information.

Location

The course will be held at Uppsala University. On Thu-Fri we will be at Geocentrum and on Mon-Tue at the University main building. See the schedule for exact locations. For information on accommodation and how to get to Uppsala, the workshop homepage might be useful.

Course level

This is a PhD level course.

Prerequisites

Basic undergraduate courses in linear algebra, probability and statistics.

Contact Persons

and Fredrik Lindsten.

Updated  2019-06-14 10:01:01 by Andreas Lindholm.