Fig. 6From: Task offloading in hybrid-decision-based multi-cloud computing network: a cooperative multi-agent deep reinforcement learningPerformance gap between the proposed CMATD3 and the theoretical optimal resultBack to article page