RL in Challenging State Spaces and Sparse Reward Signals
Talk on AlphaGo and Hierarchical RL
Introduction
This is a talk I gave at the Berlin Machine Learning Meetup on Nov 6, 2017. It is somewhat more high level than usual since the audience came from a broad range of backgrounds.
The content of the talk were ‘AlphaGo’, it’s successor ‘AlphaGo Zero’ and ‘Feudal Networks’ all published by DeepMind. The talk can be downloaded here