[PhD and Postdoc positions] Intrinsically motivated RL,
Developmental robotics, ICub
pierre-yves.oudeyer at inria.fr
Fri Jun 25 12:28:28 CEST 2010
In the context of a 3 years project funded by the french research agency ANR, we are looking for highly skilled and motivated candidates for a PhD or a Postdoc in the fields of developmental robotics, machine learning and computer vision with evaluation on the iCub humanoid robot.
The global project aims at addressing 4 complementary challenges that have been identified as crucial to the development of future humanoid robotics:
* CHALLENGE 1: How can a robot learn efficient perceptual representations of its body and of external objects given initially only low-level perceptual capabilities?
* CHALLENGE 2: How can a robot learn motor representations and use them to build basic affordant reaching and manipulation skills?
* CHALLENGE 3: What guidance heuristics, based on intrinsic motivation, active learning and social guidance, should be used to explore vast sensorimotor spaces in unknown changing bodies and environments?
* CHALLENGE 4: How can mechanisms for building efficient representations/abstractions, mechanisms for learning manipulation skills, and guidance mechanisms be integrated in the same experimental robotic architecture and reused for different robots?
These 4 interrelated challenges are distributed over four partners:
ENSTA-ParisTech (Paris) will lead challenge 1. The contact is David Filliat (david.filliat at ensta.fr).
ISIR-UPMC-Paris 6 (Paris) will lead challenge 2. ISIR hosts the iCub humanoid robot on which the acheivements will be evaluated. The contacts are Olivier Sigaud and Vincent Padois (olivier.sigaud at upmc.fr). See http://www.isir.fr/
INRIA (FLOWERS team, Bordeaux) will lead Challenge 3. The contact is Pierre-Yves Oudeyer (pierre-yves.oudeyer at inria.fr).
See http://flowers.inria.fr/ and http://www.pyoudeyer.com (for this partner, postdoc applications will be preferred)
The GOSTAI company (Paris) will lead Challenge 4. The contact is Jean-Christophe Baillie (baillie at gostai.com).
The funding should be available by January 2011 or earlier. Specific schedules might be considered for specific candidates. Salaries will depend on the candidate experience and on the partner.
The applicants should send a detailed CV and statement of motivation to the partner that is most relevant given their motivation. The topics being closely related, all candidates will be considered by all partners. Applicants should have strong competences on one or more of the following topics: developmental robotics, motor control, statistical inference, reinforcement learning, optimization, computer vision.
Detailed information about this project will be provided on request.
Don't hesitate to distribute this email to other relevant lists.
Jean-Christophe Baillie, David Filliat, Pierre-Yves Oudeyer, Vincent Padois and Olivier Sigaud
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Comp-neuro