In adaptive dynamic programming (ADP), traditional actor-critic methods make use of both policy gradient learning and value function approximation to search for near-optimal control policies in ...