AI:CogDev @ TAROS: Designing Intelligent Minds
at The University of Manchester, Manchester, United Kingdom
at The University of Manchester, Manchester, United Kingdom
Program
Keynote Speaker
Adapt to humans though mental states acquisition
Humans are constantly adapting their behaviours regarding the situations they are involved in, especially when interacting with each other. One of their main skills during social interaction is to consider the mental states of other people, such as their beliefs, preferences or intentions, and use these clues to provide suitable responses. Psychological researchers called this ability Theory of Mind (ToM) and it has become progressively more popular to adopt this advanced cognitive skill for robotic systems. Several computational models have been proposed, leveraging predictive frameworks widely used in the current literature, such as generative models and reinforcement learning methods. In this presentation, we will take a closer look at some of those models and their applications to robotic services, especially their involvement in studies of human-robot interactions.
Mehdi Hellou is a Research Fellow at the University of Manchester's Cognitive Robotics Lab (COROLAB), working within the EU's Horizon Europe programme PRIMI. Mehdi was previously an Early-Stage Researcher and PhD student at the University of Manchester (UoM), working on the Marie Sklodowska-Curie Innovative Training Networks project PERSEO, under the supervision of Prof. Angelo Cangelosi and Dr. Samuele Vinanzi.
His research focuses on improving social robot behaviours in real-world, human-centred environments by applying the premises of Theory of Mind (ToM) to enable personalised interactions. This includes designing and testing computational architectures for ToM that considers each user's individual differences and needs, tailoring the robot's actions through Reinforcement Learning techniques such as Inverse Reinforcement Learning.
Prior to his postgraduate studies at UoM, Mehdi completed a Master of Engineering in Robotics from Polytech Sorbonne University, followed by a Master's in Computer Science at Sorbonne University Pierre and Marie Curie. He also completed several internships at robotic companies, Aldebaran Robotics in Paris and robotTUNER in Wageningen.