- How to be Helpful? Implementing Supportive Behaviors for Human-Robot Collaboration (Under review). [bibtex] [pdf] [code] ,
- MCA-NMF: Multimodal concept acquisition with non-negative matrix factorization PlOS ONE, October 21, 2015. [bibtex] [pdf] [code] [data] ,
Conference and workshop proceedings
- Predicting Supportive Behaviors based on User Preferences for Human-Robot Collaboration International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2018), extended abstract. ,
- Transparent Role Assignment and Task Allocation in Human Robot Collaboration IEEE International Conference on Robotics and Automation (ICRA 2017). [bibtex] [code] [video] ,
- Learning semantic components from sub-symbolic multi-modal perception Third Joint IEEE International Conference on Development and Learning an on Epigenetic Robotics (ICDL-EpiRob 2013), Osaka (Japan). [bibtex] [code] ,
- Learning the combinatorial structure of demonstrated behaviors with inverse feedback control Human Behavior Unterstanding: Third International Workshop, HBU 2012. [bibtex] [pdf] ,
- Learning to recognize parallel combinations of human motion primitives with linguistic descriptions using non-negative matrix factorization IEEE/RSJ International Conference on Intelligent Robots and Systems, Vilamoura, Algarve (Portugal), 2012. [bibtex] [code] [data] [screencast] [details] ,
- A bag-of-features framework for incremental learning of speech invariants in unsegmented audio streams Proceeding of the Tenth International Conference on Epigenetic Robotics,, Öorenäas Slott, Sweden, 2010 (pp. 73-80). [bibtex] [pdf] ,
- The HRC Model Set for Human-Robot Collaboration Research Under review. [bibtex] [pdf] [code] ,
- The Emergence of Multimodal Concepts: From Perceptual Motion Primitives to Grounded Acoustic Words Université de Bordeaux, France, March 2014. [bibtex] [video] ,
Source code—Task models
Task models representations and algorithms for human robot collaboration.
Source code—Emergence of multimodal concept
A set of tools and experimental scripts used to achieve multimodal learning with nonnegative matrix factorization (NMF).
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.
Choreography dataset 2
This dataset is made of single gestures captured through a kinect device.
The Acorns Caregiver dataset
The Caregiver dataset was recorder by the Acorns project. It is available online thanks to Christina Bergman's work. See also Modelling the Noise-Robustness of Infants’ Word Representations: The Impact of Previous Experience for details on permissions to use the dataset. I also provide features to use directly in the multimodal experiments (see the code for more details).