Deep Reinforcement Learning for Dependent Task Offloading in Mobile Edge Computing Systems

Abstract

In mobile edge computing, there are usually relevant dependencies between different tasks, and traditional algorithms are inefficient in solving dependent task offloading problems and neglect the impact of the dynamic change of the channel on the offloading strategy. To solve the offloading problem of dependent tasks in dynamic network environment, this paper establishes the dependent task model as a directed acyclic graph. A Dependent Task Offloading Strategy (DTOS) based on deep reinforcement learning is proposed with minimizing the weighted sum of delay and energy consumption of network services as the optimization objective. The task-dependent offloading problem is transformed into an optimal policy problem under Markov decision processes. Multiple parallel deep neural networks (DNN) are used to generate offloading decisions, cache the optimal decisions for each round, and then optimize the DNN parameters using experience replay and incentive-driven mechanisms to obtain the optimal task offloading decisions. By using multiple parallel deep neural networks to generate offloading decisions, the impact of oscillations in the training results of individual neural networks on the algorithm is reduced. The experimental results show that, compared with the existing algorithms, the offloading decision generated by the algorithm can effectively reduce the delay and energy consumption of network services.