Section: Dissemination

Teaching - Supervision - Juries

Teaching

• License: Introduction à Matlab, 21 heures. L3 - ENSTA ParisTech (Alexander Gepperth)

• License: Traitement numérique du signal , 21 heures. L3 - ENSTA ParisTech (Alexander Gepperth)

• Master : Robotique de Comnpagnie, 12 heurs. M2, ENSTA - Paris Tech (Manuel Lopes).

• Master : Robotique Mobile, 24 heures. M2, ENSTA - ParisTech (David Filliat).

• Master : Vision pour la robotique, 12 heures. M2, University Pierre et Marie Curie (David Filliat).

• License : Introduction to Matlab, 21 heures. L3, ENSTA - ParisTech (David Filliat).

• License : Projet informatique, 21 heures. L3, ENSTA - ParisTech (David Filliat).

• Licence 2 : Graphe, Langage, Cryptologie, 21 heures. P™le Universitaire Francais de Ho Chi Minh Ville

• Master : Option Robotique, Projet Robot Autonome, 32 heures. ENSEIRB, Bordeaux, France.

• Licence : Mathématique, 64 heures, niveau (L2), Pôle Universitaire Français, Ho Chi Minh ville (Olivier Mangin)

• Licence : Programmation système, 30 heures, niveau (L2), Pôle Universitaire Français, Ho Chi Minh ville (Olivier Mangin)

Supervision

PhD & HdR :

• HdR: Pierre-Yves Oudeyer defended his HdR, entitled “Developmental constraints on the evolution and acquisition of sensorimotor and social skills”, at University of Bordeaux, 18 May, 2011.

• HdR : David Filliat, Navigation, Perception et Apprentissage pour la robotique, University Pierre et Marie Curie, 12 Juillet 2011.

• PhD defended by Pierre Rouanet ([21] ), “Apprendre à un robot à reconnaître des objets visuels nouveaux et à les associer à des mots nouveaux : le rôle de l'interface”, at University Bordeaux I, march 2012 (superv. Pierre-Yves Oudeyer).

• PhD in progress : Louis-Charles Caron, Developmental learning in multimodal sensory-motor loops, started january 2012 (superv. Alexander Gepperth).

• PhD in progress : Natalia Lyubova, A developmental approach to perception for a humanoid robot, started nov 2010 (superv. David Filliat).

• PhD in progress : Matthieu Lapeyre, Developmental constraints for biped humanoid walking, started oct. 2010 (superv. Pierre-Yves Oudeyer and Olivier Ly).

• PhD in progress : Mai Nguyen, Bootstrapping Intrinsically Motivated Learning with Human Demonstration, started oct. 2010 (superv. Pierre-Yves Oudeyer).

• PhD in progress : Fabien Bénureau, Cumulative, hierarchical and intrinsically motivated learning of robot skills, started oct. 2010 (superv. Pierre-Yves Oudeyer).

• PhD in progress : Jonathan Grizou, Fluid simultaneous learning of task and feedback models, started oct. 2011 (superv. Manuel Lopes and Pierre-Yves Oudeyer).

• PhD in progress : Olivier Mangin, Learning of sensorimotor primitives with Non-Negative Matrix Factorization, started oct. 2010 (superv. Pierre-Yves Oudeyer).

• PhD in progress : Thomas Cederborg, A unified view of context-dependant skill learning and language acquisition, started oct. 2009 (superv. Pierre-Yves Oudeyer).

Master internships and others:

• J. Grizou supervised the 6 month master 2 internship of Mathieu Duteil Master thesis project, coming from "Université Pierre et Marie Curie", where he studied Intelligent Systems and Robotics. The project consisted in detecting if humans provide non-verbal feedback during interactions with robots. Mathieu did several user studies to record the necessary data using several human-machine interaction protocols, recording the commands provided by the person, the sounds and the facial expressions. With this data he tested several kernel based methods to allow classifying the relevant feedback from the human.

• J. Grizou supervised two middle school students. The purpose of such one week internship is to allow pre high-school students to discover what being a researcher means as well as discovering some simple technological setup.

• F. Benureay supervised one middle school student, Victor Melançon, 16 years old, for one week internship. He worked on a visualization and control of the motion of robotic arm using a interface created during the internship using Processing. The goal was to elaborate an aesthetically pleasing visualization of the robot motion that would engage non-expert users in an interaction with the robot. The project was also the opportunity to learn the basics of programming and robot control.

• M. Nguyen supervised the internship of Thomas Huet (ENS Paris) for a 2 months internship entitled : "Learning Methods for Robotic Models in a Fishing Experiment".

• P-Y. Oudeyer supervised the 3 months master 1 internship of Sébastien Forestier (ENS Cachan). The project consisted in experimenting the SAGG-RIAC architecture on the Ergo-Robots, and it was shown to allow successful learning of inverse kinematics.