Marc Toussaint


robotics group

Machine Learning group


home
publications
software
teaching

Marc Toussaint

picture
 contact:

Marc Toussaint
TU Berlin
Franklinstr. 28/29 FR6-9
10587 Berlin, Germany
mtoussai@cs.tu-berlin.de
tel: +49 30 314 24470
room: FR 6048 (6th floor)

 secretary:

sekr@ml.cs.tu-berlin.de
tel: +49 30 314 78621
fax: +49 30 314 78622


for visitors:
map & visitor instructions

news
see our ICML 2009 papers on the publications site

see my video lectures, e.g. from ICML 2009 or UAI 2008.

see the Lectures on inference & planning at the Robot Learning Summer School 2009

source code for Stochastic Optimal Control methods (AICO, iLQG, see IMCL 2009 paper) is available here: libSOC

a new non-technical survey on the idea of using probabilistic inference for decision-making and planning

positions
current head of the Machine Learning and Robotics group (Emmy Noether Programme) at the IDA lab (Klaus-Robert Müller), TU Berlin.
8/06-2/07 guest scientist at the Honda Research Institute, Offenbach.
6/04-6/06 post doc at the Machine Learning group (Chris Williams) and the Statistical Machine Learning and Motor Control group (Sethu Vijayakumar), University of Edinburgh.
4/00-5/04 PhD student (& brief post doc) at the Adaptive Systems group, Institut für Neuroinformatik (Werner von Seelen), Ruhr-Universität-Bochum.
6/98-3/00 student at the Cologne gravity group (Friedrich W. Hehl), Institute for Theoretical Physics, U Cologne.
current research interests
  • Planning by probabilistic inference: probabilistic inference for solving (PO)MDPs, for stochastic optimal control, in robotics. Probabilistic inference as a model of goal-directed behavior in cognitive sciences.
  • Robotics: Bayesian view on motor control and planning, probabilistic models for object grasping and manipulation, motor primitives, latent variable models of motion
  • general Machine Learning: learning representations, Bayesian networks & graphical models, learning in deep factor graphs, belief propagation, structured output
  • generally, I'm also interested in relations to neuro science (e.g., the relation between neural dynamics and probabilistic inference, or between free energy models of neural dynamics and planning by inference) and to cognitive science (models of goal-directed behavior).
collaborations