ESR 14

Silvia Tulli

Socio-affective effects on the degree of human-robot cooperation

Bio

Silvia Tulli is PhD Candidate in Computer Science and Engineering at Técnico in the research group of Artificial Intelligence for People and Society (GAIPS – INESC-ID). She obtained an interdepartmental degree in Information Engineering and Computer Science and Cognitive Science from the University of Trento. She did a research period at the Institute of Science and Information Technologies (ISTI-CNR) while she worked on adaptive serious games for mild cognitive impairment and end-user development for IoT systems. She is interested in delving deeper into robotics and machine learning as disciplines of emulating living processes. Currently exploring computational learner models for robot’s plan explanations.

 

More information in silviatulli.com

Main host institution

Instituto de Engenharia de Sistemas e Computadores – Investigação e Desenvolvimento (INESC-ID)

Supervisor

Ana Paiva (INESC-ID), in association with Mohamed Chetouani (SU) and Pierre Dillenbourg (EPFL)

Second institution

Sorbonne Université ; EPFL

Objectives

There is huge potential in using social robots to help children learn more effectively in a large variety of domains. One important issue to consider is the social role that the robot should enact. Namely, it can act as a teacher that explains and presents didactic content to the children that it interacts with. Alternatively, the robot can act more as a peer or a companion that practices alongside others. It can also act as a student that children would have to teach and correct its mistakes. All these different roles have important design implications for how the robot should behave and also how the learning experience is structured. These need to be addressed properly to promote a cooperative stance between the robot and the human learner. Cultural factors will also determine the expectations regarding each different pedagogical role. This ESR project will analyse such implications and will focus on the social aspects associated to these different roles. The aim is to develop a computational model of general decision-making that allows a robot to model others from a relational perspective and adapt its expectations of cooperative behaviour accordingly. More specifically, the proposed model will enable the robot to have a better understanding of how it is perceived by learners and how appropriate are its requests during the learning tasks.

Expected results

Completed draft of PhD dissertation; implementation of computational model for robots designed to teach children, several publications in international journals and peer-reviewed conferences