Bayesian Filtering and Sequential Learning Methods

LecturerYoussef Oualil

Location:  Seminarraum (Seminar Room) C7.2  

Time: March 2017 (Block Seminar)



- The "Seminar Introduction" meeting will take place on Tuesday, February 21st, from 12:15 to 13:15 in the seminar room, building C7.2.

- Registration is now closed.

- Seminar topics were published (please see below). 



Registration is now closed: Please notice that we have a limited number of slots for this seminar. Students who could not be on the main list were moved to the waiting list. This decision is based on the registrations date/time order. 


Course Summary: In this seminar, we will review and discuss Bayesian filtering theory with a particular focus on sequential learning methods such as Kalman filter and sequential Monte Carlo approaches. We will also investigate different applications of these methods to signal and natural language processing tasks.

Seminar Format and Grading: Each of the participants will be assigned a paper or a book chapter (based on his/her preferences) that he/she needs to study and present to the rest of the group for a 50% of the grade. The remaining 50% is evaluated based on the final report following the seminar.

Seminar Topics and Dates: The topic assignment and the exact dates of the seminar will be decided based on the students preferences on different doodle polls.


The topics that will be presented during this seminar are:

1) A Closed-Form Solution for the PHD Filter.

2) An introduction to Monte Carlo Methods. 

3) An introduction to Sequential Monte Carlo Methods 

4) A tutorial on Particle Filters.

5) Bayesian Filtering with Random Finite Set Observations. 

6) Filtering via Simulation Auxiliary Particle Filters. 

7) Gaussian Mixture Sigma-Point Particle Filters for Sequential Probabilistic Inference in Dynamic State-Space Models. 

8) Joint Detection And Estimation Of Multiple Objects From Image Observations. 

9) Multiple-Hypothesis Extended Particle Filter for Acoustic Source Localization in Reverberant Environments.

10) New Developments in State Estimation for Nonlinear Systems. 

11) Recursive Bayesian Estimation-Bearings-only Applications. 

12)  Resampling Algortihms - A computational Overview. 

13) Sequential Importance Sampling Filtering for Target Tracking Image Sequences

14) Sigma-Point Kalman Filters for Probabilistic Inference in Dynamic State-Space Models.

15) Simultaneous Speaker Tracking and Microphone Array Mapping. 

16) The Gaussian Mixture PHD Filter.

17) The Split and Merge Unscented Gaussian Mixture Filter. 

18) Tracking of Multiple Speakers with Probabilistic Data Association Filters. 

19) Unscented Filtering and Nonlinear Estimation.

The topic assignment will be decided based on the students preferences using doodle.


Useful Links:


For more information, please send an email to Please include the course tag "[SLS]" in all emails related to this course.