Local Policy Optimization for Trajectory-Centric Reinforcement Learning | IEEE Conference Publication | IEEE Xplore