AAAI
2008
14 years 2 months ago
2008
Off-policy reinforcement learning is aimed at efficiently reusing data samples gathered in the past, which is an essential problem for physically grounded AI as experiments are us...