Photo: Morteza Dianatfar
In his doctoral dissertation, MSc Quentin Houbre explored how robots can autonomously discover and learn new skills by modelling key aspects of human cognition, especially curiosity. By drawing inspiration from how children learn, his work shows how perception, attention, memory, and learning can be integrated into robots to enable continuous skill acquisition. The results demonstrate that curiosity-driven learning allows robots to build predictive models of their environment and adapt their behaviour under uncertainty. Overall, Houbre suggests that developmental, curiosity-based approaches could significantly improve how robots and AI systems are trained in the future.
The doctoral dissertation of MSc Quentin Houbre in the field of Automation Science and Engineering titled “Towards active exploration and flexible goal learning: modeling curiosity as a distributed cognitive process for open-ended learning in robotics” will be publicly examined on at the Faculty of Engineering and Natural Sciences at Tampere University.
The Opponents will be Associate Professor Jörg Conradt from the Royal Institute of Technology in Stockholm (KTH) and Assistant Professor Murat Kirtay from Tillburg University (Netherlands). The Custos will be Professor Roel Pieters from the faculty of Engineering and Natural Sciences at Tampere University.