PhD students

Here is the list of PhD student I have supervised or co-supervised, together with their PhD topics, selected articles and links to their work:

Clément Romac (ongoing, co-supervised with Thomas Wolf), web: https://clementromac.github.io

Grgur Kovac (ongoing, co-supervised with Peter Ford Dominey), web: https://scholar.google.com/citations?user=ZLA7iioAAAAJ&hl=en

Thomas Carta (ongoing, co-supervised with Olivier Sigaud and Sylvain Lamprier), web: https://scholar.google.com/citations?user=-Vi3OBsAAAAJ&hl=fr

Rania Abdelghani (ongoing, co-supervised with Hélène Sauzéon and Catherine de Vulpuillères), web: https://www.linkedin.com/in/abdelghanir-rania/?originalSubdomain=fr

Laetitia Teodorescu (ongoing, co-supervised with Katja Hoffmann)

Mayalen Etcheverry (ongoing, co-supervised with Clément Moulin-Frier), web: https://mayalenetcheverry.com

Maxime Adolphe (ongoing, co-supervised with Hélène Sauzéon), web: https://www.linkedin.com/in/maxime-adolphe-b4bb96178?originalSubdomain=fr

Tristan Karch (ongoing, co-supervised with Clément Moulin-Frier), web: https://tristankarch.com

Sébastien Forestier (2019) Computational models of hierarchical curiosity-driven development of tool use and speech

  1. Web: http://sforestier.com
  2. Selected publications:
    1. Curiosity-driven Development of Tool-Use Precursors: a Computational Model
      Sebastien Forestier and Pierre-Yves Oudeyer (2016)
      Proceedings of the 37th Annual Meeting of the Cognitive Science Society, Philadelphia, USA: Cognitive Science Society.

William Schueller (2018) Computational study of the impact of active learning and active teaching in models of the formation of linguistic conventions

  1. Web: http://wschuell.github.io
  2. Selected publications:
    1. Active Learning and Active Control of Growth Complexity in Naming Games
      William Schueller and Pierre-Yves Oudeyer (2015)
      5th International Conference on Development and Learning and Epigenetic Robotics, Providence, USA. [code][Jupyter notebooks for reproducing the graphs]

Benjamin Clément (2018) Active teaching and multi-armed bandits for intelligent tutoring systems (Co-supervised with Manuel Lopes)

  1. Web: https://github.com/bclement14
  2. Selected publications:
    1. Multi-Armed Bandits for Intelligent Tutoring Systems
      Benjamin Clement, Didier Roy, Pierre-Yves Oudeyer, Manuel Lopes (2015)
      Journal of Educational Data Mining (JEDM), Vol 7, No 2.

Fabien Benureau (defended in 2014) Autonomous exploration and transfer learning in sensorimotor skill acquisition

  1. Web: http://fabien.benureau.com
  2. Current position: postdoc in Inria Mnemosyne team)
  3. Selected publications:
    1. Behavioral Diversity Generation in Autonomous Exploration Through Reuse of Past Experience
      F. Benureau, P-Y. Oudeyer (2016)
      Frontiers in Robotics and AI, 2016. [code]
    2. Diversity-Driven Selection of Exploration Strategies in Multi-Armed Bandits
      F. Benureau, P-Y. Oudeyer
      ICDL-Epirob conference, Providence, RI, USA, 2015. {pdf} {bibtex} {code} {raw plots}
  4. PhD manuscript: F. Benureau, Self-Exploration of Sensorimotor Spaces in Robots Ph.D. thesis, University of Bordeaux, May 2015. {pdf} {code}

Olivier Mangin (defended in 2014) The Emergence of Multimodal Concepts: From Perceptual Motion Primitives to Grounded Acoustic Words

  1. Web: http://olivier.mangin.com
  2. Current position: postdoc at Scassellati‘s social robotics laboratory at Yale university, US.
  3. Awards: Best PhD thesis poster — Bordeaux doctoral school of mathematics and computer science
  4. Selected publications:
    1. MCA-NMF: Multimodal concept acquisition with non-negative matrix factorization
      Mangin O., Filliat D., ten Bosch L, Oudeyer P.Y.,
      PlOS ONE, October 21, 2015.[bibtex][pdf][code][data]
    2. Learning the combinatorial structure of demonstrated behaviors with inverse feedback control
      Mangin O., Oudeyer P.Y.,
      Human Behavior Unterstanding: Third International Workshop, HBU 2012, Vilamoura, Algarve (Portugal). LNCS 7559[bibtex]
    3. Learning to recognize parallel combinations of human motion primitives with linguistic descriptions using non-negative matrix factorization
      Mangin O., Oudeyer P.Y.,
      IEEE/RSJ International Conference on Intelligent Robots and Systems, Vilamoura, Algarve (Portugal), 2012. [bibtex][data][details][screencast]
  5. Code: emergence of multimodal concept [github] A set of tools and experimental scripts used to achieve multimodal learning with nonnegative matrix factorization (NMF).
  6. Datasets: Choreography dataset [doc] [download] [see examples] This dataset is made of choreography motions captured through a kinect device. The choreography motions have a particular combinatorial structure: choreographies are designed as simultaneous execution of some primitive motions from a given set of primitive dance motions.
  7. PhD manuscript: Mangin O., The Emergence of Multimodal Concepts: From Perceptual Motion Primitives to Grounded Acoustic Words March 2014, Université de Bordeaux, France. [bibtex] [defense video]

Thomas Cederbog (defended in 2014) A formal approach to social learning: Exploring language acquisition through imitation

  1. Current position: postdoc at Charles Isbell’s AI lab, Georgia Tech, US.
  2. Award: Best PhD award in Region Aquitaine, awarded at NOVAQT science and technology festival, 2014.
  3. Selected publications:
    1. A social learning formalism for learners trying to figure out what a teacher wants them to do
      Cederborg,, T. and Oudeyer, P-Y. (2014)
      Paladyn Journal of Behavioral Robotics, 5:64-99.
    2. From Language to Motor Gavagai: Unified Imitation Learning of Multiple Linguistic and Non-linguistic Sensorimotor Skills
      Cederborg, T. and Oudeyer, P-Y. (2013)
      IEEE Transactions on Autonomous Mental Development, pp. 222-239, 5(3).
    3. Imitating Operations On Internal Cognitive Structures for Language Aquisition
      Thomas Cederborg, Pierre-Yves Oudeyer
      IEEE RAS HUMANOIDS conference, 2011, Bled, Slovenia. pp.NA, 2011
    4. Simultaneous Acquisition of Task and Feedback Models
      Manuel Lopes, Thomas Cederborg, Pierre-Yves Oudeyer
      Proceedings of Development and Learning (ICDL), 2011 IEEE International Conference on, 2011, Germany. pp.1 – 7, 2011.
    5. Incremental Local Online Gaussian Mixture Regression for Imitation Learning of Multiple Tasks
      Cederborg, T., Ming, L., Baranes, A., Oudeyer, P-Y. (2010)
      in Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010), Taipei, Taiwan.
  4. PhD manuscript: Cederborg, T. (2013) A Formal Approach to Social Learning: Exploring Language Acquisition Through Imitation, Université Sciences et Technologies – Bordeaux I, 2013. Français, Video of Talk of PhD defense

Matthieu Lapeyre (defended in 2014) Poppy: open-source 3D printed and fully modular robotics platform for science, art and education

  1. Web: https://www.linkedin.com/in/matthieulapeyre and http://www.poppy-project.org
  2. Current position: engineer at Inria, co-founder of Pollen Robotics SAS
  3. Awards: 2nd Best PhD thesis in Robotics, GdR CNRS Robotique (national prize for PhD theses in robotics in France)
  4. Selected publications:
    1. Rapid morphological exploration with the Poppy humanoid platform
      Matthieu Lapeyre, Steve N’Guyen, Alexandre Le Falher, Pierre-Yves Oudeyer, 2014
      IEEE-RAS International Conference on Humanoid Robots      , Nov 2014, Madrid, Spain. pp.8
    2. Poppy Project: Open-Source Fabrication of 3D Printed Humanoid Robot for Science, Education and Art
      Matthieu Lapeyre, Pierre Rouanet, Jonathan Grizou, Steve Nguyen, Fabien Depraetre, Alexandre Le Falher, Pierre-Yves Oudeyer
      Digital Intelligence 2014, Sep 2014, Nantes, France. pp.6, 2014
    3. Poppy Humanoid Platform: Experimental Evaluation of the Role of a Bio-inspired Thigh Shape
      Matthieu Lapeyre, Pierre Rouanet, Pierre-Yves Oudeyer
      IEEE RAS Humanoids 2013, Oct 2013, Atlanta, United States. 2013
    4. The Poppy Humanoid Robot: Leg Design for Biped Locomotion
      Matthieu Lapeyre, Pierre Rouanet, Pierre-Yves Oudeyer
      IEEE/RSJ International Conference on Intelligent Robots and Systems, Nov 2013, Tokyo, Japan. 2013
    5. Bio-inspired vertebral column, compliance and semi-passive dynamics in a lightweight robot
      Olivier Ly, Matthieu Lapeyre, Pierre-Yves Oudeyer
      IEEE/RSJ International Conference on Intelligent Robots and Systems, Jul 2011, San Francisco, United States. 2011
  5. Code/design/hardware: Design of varions open-source 3D printed robots, including Poppy humanoid robot: https://github.com/matthieu-lapeyre 
  6. PhD Manuscript: Lapeyre, M. Poppy: open-source, 3D printed and fully-modular robotic platform for science, art and education, Université de Bordeaux, 2014. Video of PhD defense ( french).

Jonathan Grizou (defended in 2014) Calibration-free interactive systems with machine learning

  1. Web: http://jgrizou.com
  2. Current position: postdoc in the Cronin group, University of Glasgow, UK (studying the properties of oil droplets as a function of their chemical composition via an automated evolutionary process.)
  3. Awards: Prize “Le Monde” for academic research (prix Le Monde de la recherche universitaire), awarded to the best 12 PhD thesis in all hard science domains in France.
  4. Selected publications:
    1. Exploiting Task Constraints for Self-Calibrated Brain-Machine Interface Control Using Error-Related Potentials
      Iñaki Iturrate, Jonathan Grizou, Omedes Jason, Pierre-Yves Oudeyer, Manuel Lopes, Luis Montesano
      PLoS ONE, Public Library of Science, 2015, [code][abstract] [bib] [latex] [pdf] [data] [plos_one] [google_scholar]

      1. Calibration-Free BCI Based Control
        Jonathan Grizou, Iñaki Iturrate, Luis Montesano, Pierre-Yves Oudeyer, Manuel Lopes
        Twenty-Eighth AAAI Conference on Artificial Intelligence, Jul 2014, Quebec, Canada. pp.1-8, 2014.
        [abstract] [bib] [latex] [pdf] [poster_latex] [poster_pdf] [google_scholar] [project_page]
    2. Interactive Learning from Unlabeled Instructions
      Jonathan Grizou, Iñaki Iturrate, Luis Montesano, Pierre-Yves Oudeyer, Manuel Lopes
      UAI-30th Conference on Uncertainty in Artificial Intelligence, Jul 2014, Quebec, Canada. pp.1-8, 2014
    3. Studying the Co-Construction of Interaction Protocols in Collaborative Tasks with Humans
      Anna-Lisa Vollmer, Jonathan Grizou, Manuel Lopes, Katharina Rohlfing, Pierre-Yves Oudeyer
      The Fourth Joint IEEE International Conference on Development and Learning and on Epigenetic Robotics, Oct 2014, Genoa, Italy
    4. Poppy Project: Open-source fabrication of 3D printed humanoid robot for science, education and art. Lapeyre, M., Rouanet, P., Grizou, J., NGuyen, S., Depraetre, F., et al. (2014). In Digital Intelligence 2014, 6–7.[abstract] [bib] [pdf] [google_scholar] [more_info]
  5. Code and robots: Libraries: [robotis-scad] [segment-scad] Robots: [poppy-ergo-jr] [poppy-4dof-arm-mini] [poppy-multipod-mini] [poppy-snake-mini] [poppy-4wheels-mini][poppy-dragster-mini]
  6. PhD Manuscript: Grizou, J. (2014) Learning from Unlabeled Interaction Frames, Univ. Bordeaux. Video of PhD defense. [abstract] [bib] [latex] [pdf] [code] [defense_slides_pdf] [defense_video] [google_scholar] [project_page]

Mai Nguyen (defended in 2014) Strategic learning through the combination of intrinsically motivated and social learning mechanisms

  1. Web: http://nguyensmai.free.fr/Home.html
  2. Current position: Maitre de conférence (associate professor, tenure position) at ENST Bretagne, France.
  3. Selected publications:
    1. Active Choice of Teachers, Learning Strategies and Goals for a Socially Guided Intrinsic Motivation Learner
      Sao Mai Nguyen, Pierre-Yves Oudeyer
      Paladyn Journal of Behavioral Robotics, Springer Verlag, 2012, 3 (3), pp.136-146. <http://link.springer.com/article/10.2478%2Fs13230-013-0110-z>. <10.2478/s13230-013-0110-z>
    2. Socially Guided Intrinsic Motivation for Robot Learning of Motor Skills
      Sao Mai Nguyen, Pierre-Yves OudeyerAutonomous Robots, Springer Verlag, 2014, 36 (3), pp.273-294. <http://link.springer.com/article/10.1007%2Fs10514-013-9339-y>.
    3. Object learning through active exploration
      Serena Ivaldi, Sao Mai Nguyen, Natalia Lyubova, Alain Droniou, Vincent Padois, David Filliat, Pierre-Yves Oudeyer, Olivier Sigaud
      IEEE Transactions on Autonomous Mental Development, IEEE, 2013, pp.1-18.
    4. Self-Organization of Early Vocal Development in Infants and Machines: The Role of Intrinsic Motivation
      Clément Moulin-Frier, Sao Mai Nguyen, Pierre-Yves
      OudeyerFrontiers in Psychology, Frontiers, 2013, 4 (1006), <http://www.frontiersin.org/Journal/10.3389/fpsyg.2013.01006/abstract>.
  4. Code: SGIM algorithm (Socially guided intrinsically motivated learning)
  5. PhD manuscript: Sao Mai Nguyen (2014) A Curious Robot Learner for Interactive Goal-Babbling : Strategically Choosing What, How, When and from Whom to Learn. Université Sciences et Technologies – Bordeaux I, 2013. English. Video/slides of presentation.

Pierre Rouanet (defended in 2012) Human-robot interfaces with touch pads as mediators object for robust teaching of new visual objects

  1. Web: https://scholar.google.fr/citations?user=UhE7ncQAAAAJ&hl=en  and http://www.poppy-project.org
  2. Current position: engineer at Inria Flowers and co-founder of Pollen Robotics SAS company
  3. Selected publications:
    1. The Impact of Human-Robot Interfaces on the Learning of Visual Objects
      Pierre Rouanet, Pierre-Yves Oudeyer, Fabien Danieau, David Filliat
      IEEE Transactions on Robotics, IEEE, 2013, 29 (2), pp.525-541. <10.1109/TRO.2012.2228134>
    2. Explauto: an open-source Python library to study autonomous exploration in developmental robotics
      Clément Moulin-Frier, Pierre Rouanet, Pierre-Yves Oudeyer
      ICDL-Epirob – International Conference on Development and Learning, Epirob, Oct 2014, Genoa, Italy. 2014
    3. Poppy Humanoid Platform: Experimental Evaluation of the Role of a Bio-inspired Thigh Shape
      Matthieu Lapeyre, Pierre Rouanet, Pierre-Yves Oudeyer
      Humanoids 2013, Oct 2013, Atlanta, United States. 2013
    4. The Poppy Humanoid Robot: Leg Design for Biped Locomotion
      Matthieu Lapeyre, Pierre Rouanet, Pierre-Yves Oudeyer
      IEEE/RSJ International Conference on Intelligent Robots and Systems, Nov 2013, Tokyo, Japan. 2013
    5. A Robotic Game to Evaluate Interfaces used to Show and Teach Visual Objects to a Robot in Real World Condition
      Pierre Rouanet, Fabien Danieau, Pierre-Yves Oudeyer
      ACM/IEEE Human-Robot Interaction HRI 2011 conference, Mar 2011, Lausanne, Switzerland. 2011
    6. Using mediator objects to easily and robustly teach visual objects to a robot.
      Pierre Rouanet, Pierre-Yves Oudeyer, David Filliat
      ACM SIGGRAPH’2010 Poster, 2010, Los Angeles, United States. 2010
  4. Code: https://github.com/pierre-rouanet
    1. Poppy project: pypot programming library, https://github.com/poppy-project/pypot
    2. Explauto: A library to study, model and simulate intrinsically motivated curiosity-driven learning and exploration in virtual and robotic agents, https://github.com/flowersteam/explauto
  5. PhD manuscript: Pierre Rouanet (2012) Apprendre à un robot à reconnaître des objets visuels nouveaux et à les associer à des mots nouveaux : le rôle de l’interface, Université Sciences et Technologies – Bordeaux I, 2012. Français

Adrien Baranes (defended in 2012) Intrinsically motivated learning and exploration: active exploration of goals and maturation for learning repertoires of sensorimotor skills

  1. Web: http://adrienbaranes.com and https://scholar.google.com/citations?user=w2I4mo4AAAAJ&hl=en
  2. Current position: Research engineer at Google, NY, US.
  3. Awards: Fullbright grant (2012), funding a postdoc at Cognitive Neuroscience lab of Jacqueline Gottlieb, Univ. Columbia, NY.
  4. Selected publications:
    1. Information-seeking, curiosity, and attention: computational and neural mechanisms.
      Jacqueline Gottlieb, Pierre-Yves Oudeyer, Manuel Lopes, Adrien Baranes
      Trends in Cognitive Sciences, Elsevier, 2013, 17 (11), pp.585-93. <http://www.sciencedirect.com/science/article/pii/S1364661313002052>.
    2. Active Learning of Inverse Models with Intrinsically Motivated Goal Exploration in Robots
      Adrien Baranes, Pierre-Yves Oudeyer
      Robotics and Autonomous Systems, Elsevier, 2013, 61 (1), pp.69-73.
    3. Eye movements reveal epistemic curiosity in human observers
      Adrien Baranes, Pierre-Yves Oudeyer, Jacqueline Gottlieb
      Vision Research, Elsevier, 2015, 117, pp.9. <http://www.sciencedirect.com/science/article/pii/S0042698915003430>.
    4. The effects of task difficulty, novelty and the size of the search space on intrinsically motivated exploration
      Adrien Baranes, Pierre-Yves Oudeyer, Jacqueline Gottlieb
      Frontiers in Neuroscience, Frontiers, 2014, 8 (317), pp.1-9. <http://journal.frontiersin.org/Journal/10.3389/fnins.2014.00317/abstract>.
    5. R-IAC: Robust intrinsically motivated exploration and active learning
      Adrien Baranes, Pierre-Yves Oudeyer
      IEEE Transactions on Autonomous Mental Development, IEEE, 2009, 1 (3), pp.155-169.
    6. Bootstrapping Intrinsically Motivated Learning with Human Demonstrations
      Sao Mai Nguyen, Adrien Baranes, Pierre-Yves Oudeyer
      IEEE International Conference on Development and Learning, 2011, Frankfurt, Germany. pp.Nguyen, 2011
    7. The Interaction of Maturational Constraints and Intrinsic Motivations in Active Motor Development
      Adrien Baranes, Pierre-Yves Oudeyer
      Proceedings of ICDL – EpiRob, Aug 2011, Frankfurt, Germany. 2011
  5. Code: Exploration Explorer for Matlab (library for studying algorithms for intrinsically motivated learning of sensorimotor skills in continuous action spaces), documentation.
  6. PhD manuscript: Adrien Baranes (2011) Motivations Intrinsèques et Contraintes Maturationnelles pour l’Apprentissage Sensorimoteur, Thèse de Doctorat de Adrien Baranès. 2011