Offloading dependent tasks in multi-access edge computing: A multi-objective reinforcement learning approach

Fuhong Song, Huanlai Xing, Xinhan Wang, Shouxi Luo, Penglin Dai, Ke Li

Abstract: Highlights•Formulate a multi-objective optimization problem of offloading dependent tasks (ODT) with dynamic user preferences.•Model a multi-objective Markov decision process with a vector-valued reward for the ODT problem.•Adapt multi-objective reinforcement learning algorithm with dynamic weight setting to the problem above.•The simulation results show that the proposed algorithm can obtain promising tradeoffs between objectives.