In the domain of robotic tutors, personalised tutoring has started to receive scientists' attention, but is still relatively underexplored. Previous work using reinforcement learning (RL) has addressed personalised tutoring from the perspective of affective policy learning. However, little is known about the effects of robot behaviour personalisation on user's task performance. Moreover, it is also unclear if and when personalisation may be more beneficial than a robot that adapts to its users and the context of the interaction without personalising its behaviour. In this paper we build on previous work on affective policy learning that used RL to learn what robot's supportive behaviours are preferred by users in an educational scenario. We build a RL framework for personalisation that allows a robot to select verbal supportive behaviours to maximise the user's task progress and positive reactions in a learning scenario where a Pepper robot acts as a tutor and helps people to learn how to solve grid-based logic puzzles. A between-subjects design user study showed that participants were more efficient at solving logic puzzles and preferred a robot that exhibits more varied behaviours compared with a robot that personalises its behaviour by converging on a specific one over time. We discuss insights on negative effects of personalisation and report lessons learned together with design implications for personalised robots.

In socially assistive robotics, an important research area is the development of adaptation techniques and their effect on human-robot interaction. We present a meta-learning based policy gradient method for addressing the problem of adaptation in human-robot interaction and also investigate its role as a mechanism for trust modelling. By building an escape room scenario in mixed reality with a robot, we test our hypothesis that bi-directional trust can be influenced by different adaptation algorithms. We found that our proposed model increased the perceived trustworthiness of the robot and influenced the dynamics of gaining human's trust. Additionally, participants evaluated that the robot perceived them as more trustworthy during the interactions with the meta-learning based adaptation compared to the previously studied statistical adaptation model.

In this paper, we investigate the applicability of deep learning methods to adapt and predict comfortable human-robot proxemics. Proposing a network architecture, we experiment with three different layer configurations, obtaining three different end-to-end trainable models. Using these, we compare their predictive performances on data obtained during a human-robot interaction study. We find that our long short-term memory based model outperforms a gated recurrent unit based model and a feed-forward model. Further, we demonstrate how the created model can be used to create customized comfort zones that can help create a personalized experience for individual users.

In this paper, we present an exploratory study of the use of tangible implicit probes to gauge the user's social engagement with a robot. Our results show that users' paying attention to the robot's implicit probes is related to higher social engagement, but also that introducing implicit probes can lead to a more positive interaction with a robot. As we observed that users in our study started paying more attention to the implicit probes after they had encountered them, the need for careful design to capture changes in social engagement over time is justified here. Finally, we discuss some of the user recommendations to design better implicit probes.