In order to successfully cooperate with humans, robots need to learn new skills and behaviours from them, for example, through gesture and speech. This type of social learning, achieved in a social context, may be facilitated by humans acting as teachers, in an implicit or explicit manner. The challenge here is to develop new statistical learning methods for social learning that find an optimal level of human intervention in the robot learning process. By adopting a breadth-first, holistic approach that integrates interdisciplinary research on social robotics and machine learning grounded in principles from the social sciences, the objective of this research is to develop computational social abilities that allow robots to behave in a socially intelligent way.
Yuan, G., Barendregt, W., Obaid, M., & Castellano, G. (2018). When robot personalisation does not help: Insights from a robot-supported learning study. IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN 2018). [PDF]
Sequeira, P., Alves-Oliveira, P., Ribeiro, T., Di Tullio, E., Petisca., S., Melo, F. S., Castellano, G., & Paiva, A. (2016). Discovering Social Interaction Strategies for Robots from Restricted-Perception Wizard-of-Oz Studies. In Proceedings of the 2016 ACM/IEEE International Conference on Human-Robot Interaction (HRI 2016).Best paper award. [PDF]
Leite, I., Castellano, G., Pereira, A., Martinho, C., & Paiva, A. (2014). Empathic Robots for Long-term Interaction: Evaluating Social Presence, Engagement and Perceived Support in Children. International Journal of Social Robotics, 6(3): 329-341. [PDF]