A Novel Reinforcement-Based Paradigm for Children to Teach the Humanoid Kaspar Robot

Abstract

This paper presents a contribution aiming at testing novel child-robot teaching schemes that could be used in future studies to support the development of social and collaborative skills of children with autism spectrum disorders (ASD). We present a novel experiment where the classical roles are reversed: in this scenario the children are the teachers providing positive or negative reinforcement to the Kaspar robot in order for it to learn arbitrary associations between different toy names and the locations where they are positioned. The objective is to stimulate interaction and collaboration between children while teaching the robot, and also provide them tangible examples to understand that sometimes learning requires several repetitions. To facilitate this game, we developed a reinforcement learning algorithm enabling Kaspar to verbally convey its level of uncertainty during the learning process, so as to better inform the children about the reasons behind its successes and failures. Overall, 30 typically developing (TD) children aged between 7 and 8 (19 girls, 11 boys) and 9 children with ASD performed 25 sessions (16 for TD; 9 for ASD) of the experiment in groups, and managed to teach Kaspar all associations in 2 to 7 trials. During the course of study Kaspar only made rare unexpected associations (2 perseverative errors and 2 win-shifts, within a total of 314 trials), primarily due to exploratory choices, and eventually reached minimal uncertainty. Thus, the robot's behaviour was clear and consistent for the children, who all expressed enthusiasm in the experiment.

Artificial Intelligence Enabled Personalised Assistive Tools to Enhance Education of Children with Neurodevelopmental Disorders—A Review

“…Additionally, the application has enabled participants to experience more engagement and exhibit higher behavioural intentions towards it. A study that used 'Facesay' games to aid autistic students [86] reported improvements in emotion recognition, social interaction, facial recognition, emotion recognition and social interaction in low-functioning and high-functioning autistic students. The application has also been found to be very promising, cost-effective and efficient for teaching effect recognition and mentalising constructs to high-functioning ASD students.…”

“…The development of software games to help students with challenging behaviour of ADHD students has resulted in positive impacts [85] in addressing their moods, wherein a reduction in challenging behaviours of participants was reported when the games were used as an intervention to improve behaviours. Apart from the applications, the use of robots [84,85,92,93,102,103] and alternative communication devices [78,86] have been proven to be effective in improving focus, as well as math and social skills and in the teaching of ASD students, respectively. Hence, the abovementioned findings confirm the effectiveness of AI tools for personalised education.…”

Kaspar in the wild: Experiences from deploying a small humanoid robot in a nursery school for children with autism

“…While those studies with Kaspar were conducted independently of our research team at the University of Hertfordshire, our team has also conducted a number of other more controlled studies in schools. This includes a study using the robot in a different cultural context, namely, in a Greek school [76] as well as studies investigating how Kaspar can teach children with autism about visual perspective taking, for examples see [29], and how the robot can learn from interactions with children with autism [77]. Moreover, Kaspar was used independently by a clinician in the University Children's Hospital-Skopje, Macedonia, and case studies with children with severe autism are provided in [78].…”

From Humans and Back: a Survey on Using Machine Learning to both Socially Perceive Humans and Explain to Them Robot Behaviours

“…Without having studies which understand how human beings interpret the behaviour of robotic systems and what are their expectations of such systems, it would be impossible, for example, to help humans trust their robot counterparts in critical and non-critical situations [15]. On this subject, Han et al [16] proposed a literature review, while interesting studies from a computer science, artificial intelligence, cognitive psychology, and philosophical [29] Automated rationale generation Recurrent neural networks [26] Goal communication, favourizing user anticipation Inverse reinforcement learning [63] Training personalized policies for teaching Model-free affective reinforcement learning Tega [67] Teaching a robot to learn toy names and the locations Reinforcement learning Kasper [42] Robot approaching behaviour to groups of humans Proximal policy optimization Pepper [35] Adaptively decide a monitoring distance and an approaching direction to improve user activity recognition performance perspective can be found in a recent full-day workshop [17]. For example, the Theory of Mind (ToM) model is used by both human and robot in order to understand each other in an autonomous car driving on the highway scenario [18].…”

“…In a similar educational scenario, decision trees, neural networks, and clustering algorithms are used to learn from social interactions [64]. Other machine learning approaches have also been used in learning by teaching scenarios with children, notably principal component analysis (PCA) [65], and inverse optimal control [66] to model handwriting and RL to learn toy's names and locations [67].…”