Accelerating Reinforcement Learning

Ljubisa Basic and Professor Matt Taylor discuss the role of Reinforcement Learning on the path to general intelligence. Overview:
Introduction
Reinforcement learning: overview 2:05
Core RL questions 6:15
Recent RL achievements 11:49
Components of RL systems 15:22
Hardware limitations and RL 23:38
Parallel computing 26:31
Sparsity, latency and batch size 41:08
The next big thing in RL 43:13