Main

Home Page

[ . . .w.e.l.c.o.m.e. . . . .b.i.e.n.v.e.n.u.e. . . ]

Shortcuts / Raccourcis:

http://www.sauvonsluniversite.com/

Update: Proceedings for the Workshop on Evolutionary Robotics at IEEE IROS 2009 (Saint Louis, USA, october 2009) are available

[http://www.isir.fr/evoderob]

Hello, my name is Nicolas Bredeche, I am Maitre de Conférences HDR (i.e. ~Assistant Professor, with habilitation thesis) at the Université Paris-Sud XI (France, 21km from Paris with a bicycle). I am a researcher within the Machine Learning and Evolutionary Computation team (Equipe Inférence et Apprentissage, équipe-projet TAO, INRIA Saclay) at the Laboratoire de Recherche en Informatique (LRI/CNRS, Univ. Paris-Sud). I am currently delegated as a full time INRIA researcher in the TAO team (delegation INRIA, 2008-2009).

  • main theme: Artificial Intelligence / Intelligence Artificielle
  • sub-themes:
    • developmental design (multi-cellular model, amorphous computing, artificial ontogeny)
    • evolutionary robotics (neuro-control, swarm robotics, artificial life)
  • in a nutshell: I'm interested in optimizing and studying the dynamics of complex systems such as swarm of robots or assemblage of active modules, would it be either for discovering new and original behaviors and/or organization (e.g. how a swarm of robots may evolve), or to search for the optimal behavior for solving some user-defined problem (e.g. how to best explore a new environment, how to localize a group of agents, how to maintain stability for a construction of active blocks).
  • en deux mots: mon travail s'inspire (de loin) de l'évolution naturelle pour optimiser et étudier des systèmes distribués composés de plusieurs agents autonomes tels que des essaims de robots ou des assemblages de modules actifs, que ce soit pour découvrir des comportements nouveaux et originaux (p.ex. comment s'organise une société de robots), ou pour trouver des solutions à des problèmes posés par un superviseur humain (p.ex. comment un robot peut-il explorer une zone au mieux, comment un groupe de robots peut-il se localiser dans un environnement inconnu, comment une construction composée de petites unités "intelligentes" peuvent elles agir pour que l'ensemble reste stable).

As a researcher, i am concerned with the automatic conception through artificial evolution of agent behaviors, either single autonomous agent (such as virtual or real robot) or group of autonomous agents (such robot swarm). Evolutionary Algorithms (EA) are at the core of my research as they can be either considered as (1) a specific flavor of stochastic optimization algorithms from the optimization toolbox which are particularly fitted when the fitness function is poorly informative or even ill-defined (optimization issues) or (2) a very handy discovery algorithm that can provide insights on new kind of solutions for an open problem (discovery issues).

In this scope, i am concerned with issues dealing with two complementary domains: that of evolving/learning neuro-controllers (expressivity, learnability, evolvability and scaling issues) and that of evolving complex systems where each element has only a limited perception/knowledge of the environment/problem (synchronisation, self-stabilisation, self-repairing, robustness and scaling (again) issues). Examples of applications i am concerned with include behavior optimization for swarm of robots (e.g. European Project Symbrion (IP, FP7)) and artificial embryogeny approach for evolving stable and self-reparing designs such as "active" metallic trusses or continuous CA (e.g. solving stability issues in metallic truss through a distributed approach). You can refer to the publications section for further information (papers and co-authors).

Some illustrations from research works i am (or was) involved in. (see Introduction and/or Publications sections for references and collaborators).

:SHORT BIO:

I graduated in Computer Science from the University of Paris 6 (1998) and obtained a Master's degree in Cognitive Science from the University of Paris-Sud XI in 1999. I pursued a Ph.D. (1999-2002, Machine Learning and Robotics) in the Computer Science Lab. of University of Paris 6 (LIP6) under the supervision of Pr. J.-D. Zucker and Pr. G. Sabah. Then, I worked as a post-doc researcher at ICT/CAS (2003, Chinese Academy of Science, Beijing, China) under the supervision of Pr. S. Zhongzhi. In september 2003, I was recruited as assistant professor at the University of Paris-Sud XI and i've been working as a researcher in the IA-TAO team since then. In sept. 2008, I was lucky enough to be awarded a delegation from INRIA and I'll be working as a full time researcher at INRIA Saclay in the TAO team for two years (delegation INRIA, 2008-2010).

2003-present : Maitre de Conférences / Assistant Professor (University Paris-Sud 11)

  • Research: Optimization and Machine Learning team (TAO) ; LRI/CNRS, INRIA, University Paris-Sud.
  • Teaching: Université Paris-Sud XI, IFIPS engineering school.
  • Visiting Scientist:
    • Computational Intelligence Group at Vrije Universiteit in Amsterdam, for three months (Jan.-April 2009) in the scope of the Symbrion project.
    • EVO-DesignOpt group at CSAIL, MIT (Massachusetts Institute of Technology) for two weeks in Nov. 2007 and one week in July 2009.
    • MICA/CNRS (Hanoi, Vietnam) for one month in Aug./Sept. 2007.
  • 2008-2010: temporary position as INRIA researcher for two years ("Delegation INRIA").
  • 2009 (december): Habilitation Thesis defense (HDR - Habilitation à Diriger des Recherches) entitled "Contributions to Evolutionary Design of Embodied Agents" (Univ. Paris-Sud)
    • Jury: Una-May O'Reilly (MIT) reviewer ; Francois Charpillet (INRIA LORIA) reviewer ; Dario Floreano (EPFL) reviewer ; A.E. Eiben (VU Amsterdam) ; Wolfgang Banzhaf (Univ. of Newfoundland) ; Philippe Bidaud (Univ. Paris 6) ; Marc Schoenauer (INRIA)

2003 : Post-doc at ICT/CAS (China)

  • Institute of Computing Technology, Chinese Academy of Sciences (Beijing, China)
  • From january to september 2003
  • Supported by a French Lavoisier Research Grant.

1999-2002 : Doctorat / Phd in Computer from University Paris-Sud XI.

  • Phd in Artificial Intelligence (advisors: Jean-Daniel Zucker (now senior researcher at IRD) and Gerard Sabah (senior researcher at CNRS).
  • Physical location : LIP6/CNRS, Université Pierre et Marie Curie - Paris 6, France.
  • Defended on december 12th, 2002. Jury: Jean-Daniel Zucker (advisor), Gérard Sabah (co-advisor), Lorenza Saitta (reviewer), Luc Steels (reviewer), Alexis Drogoul (examiner), Jean-Sylvain Liénard (examiner and president of jury).
  • Keywords: Symbol Grounding, Machine Learning, Feature Construction, Object Identification for Autonomous Robots.

1998-1999 : DEA in Cognitive Science (Master 2)

  • Courses location: LIMSI/CNRS, Université Paris-Sud XI.
  • Internship: LIP6/CNRS, Université Pierre et Marie Curie - Paris 6, France.

To contact me or to visit me, please click on the contact link on the menu bar. You may also click here.