Abstract
We show that several popular discounted reward natural actor-critics, including the popular NACLSTD and eNAC algorithms, do not generate unbiased estimates of the natural policy gradient as claimed. We derive the first unbiased discounted reward natural actor-critics using batch and iterative approaches to gradient estimation. We argue that the bias makes the existing algorithms more appropriate for the average reward setting. We also show that, when Sarsa(λ) is guaranteed to converge to an optimal policy, the objective function used by natural actor-critics has only global optima, so policy gradient methods are guaranteed to converge to globally optimal policies as well.