Vous êtes ici : Accueil » Kiosque » Annonce

Identification

Identifiant: 
Mot de passe : 

Mot de passe oublié ?
Détails d'identification oubliés ?

Annonce

11 mars 2021

Stage Apprentissage robotique : Continuously Learning Complex Tasks via Symbolic Analysis


Catégorie : Stagiaire


Ecole Polytechnique & Ensta have a 6 months internship position at the intersection of Robotics and Formal Methods.
The goal is to explore set-based and symbolic reasoning to tackle the challenges of lifelong learning of hierarchical tasks in DeepLearning.

Summary:

Fully autonomous robots have the potential to impact real-life applications, like assisting elderly people. Autonomous robots must deal with uncertain and continuously changing environments, where it is not possible to program the robot tasks. Instead, the robot must continuously learn new tasks. The robot should further learn how to perform more complex tasks combining simpler ones (i.e., a task hierarchy). This problem is called lifelong learning of hierarchical tasks.
The existing learning algorithm for hierarchical tasks are limited in that: a) they require the robot to execute a large number of real actions to sample the continuous state space of observations, hence requiring a lot of time; b) they cannot deal with subspaces without continuous interpolation, as it is the case for a hierarchy of tasks.
In this internship, we will exploit the intuition that set-based reasoning of the continuous space can reduce the number of samples required to learn a hierarchy of tasks and allow for more effective planning of the robot tasks, further handling discontinuities in the task hierarchies. This internship will analyse the state of the art and compare existing approaches. It will also seek and design a benchmark to test the ability of algorithms to learn hierarchical tasks.

 



Required Skills for the Candidate:
We are looking for a student with a Master degree in Computer Science. The ideal candidate will have a Master degree in Computer science and a strong background in at least one topic among learning algorithms, planning, and formal methods (e.g., abstract interpretation, model checking). The ideal candidate will then acquire the required knowledge in robotics.

More more details, please contact :
• Sergio Mover (sergio.mover@lix.polytechnique.fr), Cosynus, Laboratoire d’informatique de École polytechnique (LIX), École Polytechnique
• Sao Mai Nguyen (nguyensmai@gmail.com), Autonomous Systems and Robotics team, Computer Science and Systems Engineering Laboratory (U2IS), ENSTA Paris

References
[1] Y. Dudai. The Restless Engram: Consolidations Never End. In Annual Review of Neuroscience, volume 35. 2012. [2] A. Manoury, S. M. Nguyen, and C. Buche. Hierarchical affordance discovery using intrinsic motivation. In HAI,
2019.
[3] Sebastian Thrun and Lorien Pratt. Learning to Learn. Springer US, Boston, MA, 1998.

 

Dans cette rubrique

(c) GdR 720 ISIS - CNRS - 2011-2020.