Dr. Pierre-Yves Oudeyer is Research Director (DR1) at Inria and head of the Inria and Ensta-ParisTech FLOWERS team (France). He studies computational mechanisms allowing robots and humans to acquire open-ended repertoires of skills through lifelong learning. This includes natural and artificial intelligence processes for curiosity, intrinsic motivation, the role of morphology in learning motor control, human-robot interfaces, joint attention and joint intentional understanding, and imitation learning. He has published a book, more than 100 papers in international journals and conferences, holds 8 patents, gave several invited keynote lectures in international conferences, and received several prizes for his work in developmental robotics and on the origins of language. Web: http://www.pyoudeyer.com and http://flowers.inria.fr

Intrinsically motivated exploration is a key mechanism for autonomous lifelong learning in human children.

This allows them to discover and acquire large repertoires of skills, through self-generation, self-selection, self-ordering and self-experimentation of learning goals.

I will present a formal framework to conceptualize various forms of intrinsically motivation exploration in humans and machines.

In particular, I will present an algorithmic approach called intrinsically motivated goal exploration processes, and show through various

experimental setups how it can allow robots to discover and learn repertoires of diverse high-dimensional skills, ranging from the manipulation

of soft objects to the discovery of nested tool use. I will also show how these mechanisms generate automatically a learning curriculum.

While no particular target task is provided to the system, this curriculum allows the discovery of tasks of increasing complexity, that act as stepping stones

for learning more complex skills that would be otherwise very difficult to learn.