Learning Highway Ramp Merging Via Reinforcement Learning with Temporally-Extended Actions | IEEE Conference Publication | IEEE Xplore