1 paper · 1 filter
Shreyas Chaudhari, Renhao Zhang, Philip S. Thomas +1
The ability of reinforcement learning algorithms to learn effective policies is determined by the rewards available during training. However, for practical problems, obtaining larg…