Deep Reinforcement Learning techniques for dynamic task offloading in the 5G edge-cloud continuum | Publicación