Continual reinforcement learning

Intelligent agents must be able to learn by interacting with their environment and to adapt to changes. Continual reinforcement learning provides a natural way to model this process. In this talk, I will discuss my point of view regarding how we can formalize continual reinforcement learning, and the types of methods that can be used to tackle it. The ideas draw both from algorithmic reinforcement learning and from cognitive science and psychology notions such as complementary learning systems, plasticity and empowerment.