A Deep-Reinforcement-Learning-Based Approach to Dynamic eMBB/URLLC Multiplexing in 5G NR | IEEE Journals & Magazine | IEEE Xplore