The theme of this workshop is “Cumulative Learning: Acquiring competence from a lifetime of experience”.  We have seen many developments in recent years, including work on option discovery, continual learning, constructivism, hierarchical reinforcement learning, and motor skill discovery.   We have also seen impressive performance from providing more data and computation to simple reinforcement learning algorithms using deep neural networks for function approximation.  In this workshop we want to examine our state of knowledge.  Do we know enough to build agents that can acquire a diverse competencies from a single extended lifetime?  What are the opportunities and what are the potential pitfalls?

Potentially relevant sub-topics include (but are not limited to):