Fast Online Reinforcement Learning Control Using State-Space Dimensionality Reduction | IEEE Journals & Magazine | IEEE Xplore