Computation Offloading in Beyond 5G Networks: A Distributed Learning Framework and Applications

Facing the trend of merging wireless communications and multi-access edge computing (MEC), this article studies computation offloading in beyond fifth generation networks. To address technical challenges originating from uncertainties sharing limited resource an MEC system, we formulate problem as a multi-agent Markov decision process, for which distributed learning framework is proposed. We present case study on orchestration to showcase potential online reinforcement algorithm developed under proposed framework. Experimental results demonstrate that our outperforms benchmark algorithms. Furthermore, outline research directions worth in-depth investigation minimize time cost, one main practical issues prevent implementation