Section: New Results

Educational Technologies

KidLearn

Participants : Manuel Lopes [correspondant] , Pierre-Yves Oudeyer, Didier Roy, Benjamin Clement.

Kidlearn is a research project studying how machine learning can be applied to intelligent tutoring systems. It aims at developing methodologies and software which adaptively personalize sequences of learning activities to the particularities of each individual student. Our systems aim at proposing to the student the right activity at the right time, maximizing concurrently his learning progress and its motivation. In addition to contributing to the efficiency of learning and motivation, the approach is also made to reduce the time needed to design ITS systems.

Intelligent Tutoring System (ITS) are computer environments designed to guide students in their learning. Through the proposal of different activities, it provides teaching experience, guidance and feedback to improve learning. The FLOWERS team has developed several computational models of artificial curiosity and intrinsic motivation based on research on psychology that might have a great impact for ITS. Results showed that activities with intermediate levels of complexity, neither too easy nor too difficult but just a little more difficult that the current level, provide better teaching experiences. The system is based on the combination of three approaches. First, it leverages Flowers team’s recent models of computational models of artificial curiosity and intrinsic motivation based on research in psychology and neuroscience. Second, it uses state-of-the-art Multi-Arm Bandit (MAB) techniques to efficiently manage the exploration/exploitation challenge of this optimization process. Third, it leverages expert knowledge to constrain and bootstrap initial exploration of the MAB, while requiring only coarse guidance information of the expert and allowing the system to deal with didactic gaps in its knowledge. In 2014, we have run a second pilot experiment in elementary schools of Région Aquitaine, where 7-8 year old kids could learn elements of mathematics thanks to an educational software that presented the right exercises at the right time to maximize learning progress. [69] , [37] , [38] , [39]

Education and the Poppy project

Participants : Matthieu Lapeyre [correspondant] , Pierre-Yves Oudeyer, Didier Roy.

The Poppy platform was initially designed for research purposes and even more specifically for studying biped locomotion and human-robot interaction. However, it has been designed with open science goals in mind, both to share our research and create tools for researchers. As we are convinced of the need for multidisciplinary contributions in order to improve the state of the art in the robotics field, we decided right from the beginning to use and create modern and easy-to-use tools. This choice has strongly affected the way we designed our platform. Indeed, being simple to use, easily reproducible and hackable, modular, 3D printable and as plug 'n play as possible lead to the development of hardware (Poppy) and software (pypot) tools that can be also used by non-expert people.

Thus Poppy meets a growing societal need: education and training in technologies combining computer science, electronics and mechanics, as well as a training tool for the emergent revolutionary 3D printing process. Since October 2013 (open source release), we have been contacted by several Fablabs, universities, engineering schools and even high schools. We have had the opportunity to meet with educational teams and it appears they are looking for new motivational tools for group projects.

In this context, the Poppy platform appears well suited. Indeed, it integrates advanced and yet easily accessible techniques (3D printing, Arduino, Python) in an embodiment that motivates students and the wider public. With its openness, design and rather low-cost, Poppy is highly hackable and provides a unique context for learning and experimenting with these technologies in a Do-It-Yourself (DIY) way.

The paper [54] describes the use of the Poppy platform as a tool for scientific researches as well as educationnal and artistic applications.

Several experiments with Poppy in middle and high schools, science museums and Fablabs in France and abroad are already underway and will be discussed in the upcoming Partnerships and Cooperations sections.

Expression of emotions with Poppy Humanoid

Participants : Fabien Benureau [correspondant] , Matthieu Lapeyre.

Two students in 3rd year of the Cognitive Science major at the University of Bordeaux led a TER project this year using Poppy under the supervision of Fabien Benureau, exploring how the attitude towards robots influences how humans recognise the emotion they try to express. Poppy having no facial expression — or face — yet, the students expressed the five expressions they selected (anger, surprise, joy, sadness, disgust) with body movements alone. They videotaped the sequences of movements (videos are available here http://python.sm.u-bordeaux2.fr/ter/2014/sc/desprez-zerdoumi/?page_id=289 ) and created an experiment asking volunteers to guess which emotion was displayed. The form also included the Negative Attitude towards Robots Scale (NARS), to investigate the possible correlation between fear of robot and the ability to identify their emotional attitude. The results showed no correlation between the two, although it was admitted that the experiment would have to be improved and ran again before any conclusion could be made.