Developing creativity in cognitive robots

I aim to build robots capable of insight using Hierarchical Reinforcement Learning.

I am investigating the application of hierarchical reinforcement learning in the modeling of insight in robots. Insight problems require going against one's initial intuition and making use of experience to discover an original – and better – solution. In humans and animals, insight is considered a hallmark of creativity. To reproduce this in robots, several challenges must be met – in particular, the combination of learning and searching, and the automatic development of a hierarchy of policies through interaction with the world. A robot capable of meeting these challenges will solve insight problems in a manner consistent with the bulk of the psychological literature on insight.

Secondments:

Nikolas Hemion; Aldebaran robotics, Paris

Research Fellow
Thomas R. Colin
Supervisors

Tony Belpaeme, Angelo Cangelosi, Michaela Gummerum, Thomas Wennekers (Plymouth University), Nikolas Hemion (Aldebaran robotics)