Fig. 9From: Decentralized computation offloading for multi-user mobile edge computing: a deep reinforcement learning approachIllustration of the average reward in the training process with wm=0.5 for all mobile usersBack to article page