I am currently a PhD student in the Equipes Traitement de l'Information et Systèmes (ETIS) laboratory at Cergy-Pontoise, in France. I started my PhD in 2010 in the Neurocybernetic team under the supervision of Pr. Philippe GAUSSIER, Pr Philippe TARROUX and Dr. Nicolas CUPERLIER. The topic of my PhD thesis is about "From self-evaluation to emotions: neuromimetic and bayesian approaches for the learning of complex behavior involving multimodal informations". I am to graduate by the end of 2013.
About Neurocybernetics team
The research activity of the team aims at understanding the mecanisms enabling a living being to adapt to its environment (insect, animal or human) and at implementing them on autonomous robots which use vision as main source of information.
The followed approach is bottom-up as it considers that the conjunction of low level mecanisms with capabilities of adaptation and a robotic setup embedded in a rich environment can let the emergence of stable behaviors. This approach implies to consider the dynamics of the robotic system as well as the global dynamic of the interactions between the system(s) and the environment - whether physical or social. To do so, sensorimotor loops are designed as artificial neural networks with capabilities for learning (associative, unsupervised, conditionning or reinforcement learning), and for parallelization (splitting and distributing the neural loops over several computing units, reusing the same cortical and sub-cortical "structures" in several models).
Thus, the same tools are used to study the issues of motor control, multi-modality, planning and selection of action for navigation and non-verbal human-robot cooperation tasks requiring the understanding of the mecanisms of recognition, affordance, imitation and emotional interactions.
- Neural networks, bio-inspired robotics
- Sensorimotor control
- Learning from demonstration, interactive learning, imitation
- Developmental approach