Discrete-time generalized policy iteration ADP algorithm with approximation errors | IEEE Conference Publication | IEEE Xplore