COCAM: A Cooperative Video Edge Caching and Multicasting Approach Based on Multi-Agent Deep Reinforcement Learning

Abstract The evolution of the Internet of Things technology has boosted the drastic increase in network traffic demand. Caching and multicasting are effective approaches to alleviate the backhaul burden of networks and reduce service latency. However, existing works do not jointly exploit the advantages of these two approaches. In this paper, we propose COCAM, a cooperative video edge caching and multicasting approach based on multi-agent deep reinforcement learning to minimize the transmission number in the multi-clouds scenario with limited storage capacity in each mobile edge server. Specifically, by integrating a cooperative transmission model with the caching model, we provide a concrete formulation of the joint problem. Then, We cast this decision-making problem as a multi-agent extension of the Markov decision process and propose a multi-agent actor-critic algorithm in which each agent learns a local caching policy and observes the information of neighboring agents as part of the environment. Finally, to validate the COCAM algorithm, we conduct extensive experiments on a real-world dataset. The results show that our proposed algorithm outperforms other baseline algorithms in terms of the number of video transmissions.