Hierarchical Model-Based Reinforcement Learning

Hierarchical decomposition promises to help scale reinforcement learning algorithms naturally to real-world problems by exploiting their underlying structure. Model-based algorithms, which provided th