Deep reinforcement learning-based task offloading and resource allocation for mobile edge computing

Abstract

We consider a mobile edge computing system that every user has multiple tasks being offloaded to edge server via wireless networks. Our goal is to acquire a satisfactory task offloading and resource allocation decision for each user so as to minimize energy consumption and delay. In this paper, we propose a deep reinforcement learning-based approach to solve joint task offloading and resource allocation problems. Simulation results show that the proposed deep Q-learning-based algorithm can achieve near-optimal performance.