Understanding the Effects of Second-Order Approximations in Natural Policy Gradient Reinforcement Learning

Open in new window