One major difference between HDP and DHP is within the objective of the critic network. In HDP, the critic network outputs the estimate of J in (2) directly, while DHP estimates the derivative of J with respect to its input vector.
One major difference between HDP and DHP is within the objective of the criticnetwork. In HDP, the critic network outputs the estimate of J in (2) directly, while DHP estimates the derivative of J with respect to its input vector.