Learning Control Policies from Optimal Trajectories | IEEE Conference Publication | IEEE Xplore